Phase difference between Electric field computed by FEKO and using Collin formulas
Good morning, I would like to ask a question.
I need to compute the Near Field x-component of the electric field radiated by a short current filament over a plane.
In Feko the antenna is modeled as an electric dipole source
The current filament is located in the origin of an xyz leftwards system
The product of the current over the filament by the length of the antenna I * dl is set equal to 1 A*m.
lambda = wavelength is 1 m
The plane is an xy plane located at z = d = 6 * lambda
The plane is a square with side 10 * lambda centered in 0
The mesh over the plane is a 21 x 21 grid points in xy with lambda/2 spacing
I load and use the xy grid computed by Feko to grant consistency
I then compute the Ex with the formulas proposed by R.E. Collin, "Antennas and Radiowave propagation",Internation Student Edition, for a short current filament (formula 2.28) projected over the x dimension
The spherical coordinate system (r, theta, phi) in use has theta = angle between positive z axis and r, ranging from 0 to pi, phi = angle between the positive x direction and the projection of the radius on the x-y plane , ranging from 0 to 2*pi
so that
x = r * sin(theta) * cos(phi)
y = r * sin(theta) * sin(phi)
z = r * cos(theta)
Why can I observe a slight phase difference between the Ex field computed by Feko and the Ex field computed with the Collin formulas?
Attached you can find the Matlab script in use and the output figures
Thank you,
Regards,
FB
Best Answer
-
I checked your Feko/Matlab files and found out the following mistakes in your Matlab (script) file.
1) E_r_collin_plane(p,q) = 1i * Z0 * I * dl /(2 * pi * k) * cos(theta) * (1i * k /r + 1/r^2) * exp(-1i * k * r)/r;
-> E_r_collin_plane(p,q) = -1i * Z0 * I * dl /(2 * pi * k) * cos(theta) * (1i * k /r + 1/r^2) * exp(-1i * k * r)/r; <- Please add negative sign (-). I compared your equations with the original in a book.
2) d = 6 * lambda -> d =6 <- d=6 is more exact value because z = 6.
After modifying the two, the following comparisons were made by Altair Compose. Thank you.
0
Altair EmployeeAnswers
-
Could you also share your Feko .cfx and "dipole_feko_efe.txt" files? Maybe I can run your Matlab file with Altair Compose..(I don't know why I can't see your previous two figures. Please also share the two again.)
0 -
Jaehoon Kim_22316 said:
Could you also share your Feko .cfx and "dipole_feko_efe.txt" files? Maybe I can run your Matlab file with Altair Compose..(I don't know why I can't see your previous two figures. Please also share the two again.)
Thank you so much! I have also attached the original .efe output file. Its name is "prova1_dipole_NearFieldFranci_piano_d_6.efe" . The txt file is just its replica without the header .
0 -
I checked your Feko/Matlab files and found out the following mistakes in your Matlab (script) file.
1) E_r_collin_plane(p,q) = 1i * Z0 * I * dl /(2 * pi * k) * cos(theta) * (1i * k /r + 1/r^2) * exp(-1i * k * r)/r;
-> E_r_collin_plane(p,q) = -1i * Z0 * I * dl /(2 * pi * k) * cos(theta) * (1i * k /r + 1/r^2) * exp(-1i * k * r)/r; <- Please add negative sign (-). I compared your equations with the original in a book.
2) d = 6 * lambda -> d =6 <- d=6 is more exact value because z = 6.
After modifying the two, the following comparisons were made by Altair Compose. Thank you.
0 -
Thank you so much for your precise help. With your correction I was able to fix all the issues.
I have to assume that the formula on Collin is incorrect. I attach a screenshot, four eyes are always better than two.
Best Regards
0