How do I compute the current through a plane or along a surface in Feko?

tobias_from_edag · 2025-02-03T14:59:11+00:00

There was an error rendering this rich post.

Hello, I am trying to find a way to compute the total current through a small plane. Or alternatively to compute the current along a geometric object.

For example, if I have a connecting element represented by a 1D-wire, I can read the current in the currents results. I am looking for a procedure that produces the same number for the same element if it is represented by a 3D-zylinder. And ideally one that can be generalized for non-cylindrical objects.

Find more posts tagged with

Feko

v2024

Accepted answers

tobias_from_edag

Hello future Readers,

I did end up re-asking this question to support and got an answer:

One can split the conductor and add an edge port to the split point. Then adding a voltage source allows one to read the current.

They also wrote a script that finds the current using Ampere's law from a previously defined nearfield :

-- Compute common mode current vs. distance along the cable
app = pf.GetApplication()

-- Retrieve near field request. Assume cylindrical coordinates around cable
model = "EMC_BCI_DUT_2024.1"
request = "FE_Around"
feRequest = app.Models[model].Configurations[1].NearFields[request]
feData = feRequest:GetDataSet()

--Create a new (empty) result set to store commonm ode current
iSumData = pf.DataSet.New()

--Add axes to the new result set - using the frequency axis from the original nearfield
iSumData.Axes:Add(feData.Axes[1]) -- Frequency
iSumData.Axes:Add(feData.Axes[2]) -- rho
iSumData.Axes:Add(feData.Axes[4]) -- X

--Add quantities to the new result set
iSumData.Quantities:Add('Ix', 'complex', 'A') -- Ix

--Show the structure of the new dataset
print(iSumData)

-- Number of phi samples -- assume it is te same for all near-field requests
nphi = #feData.Axes[3]

-- Number of frequency samples -- assume it is te same for all near-field requests
nfreq = #feData.Axes[1]

-- Loop over the number of frequency points
for freq_count = 1, nfreq do
rho = 0
for u_count = 1, #feData.Axes[2] do
rho = feData.Axes[2][u_count]
-- print(rho)
for n_count = 1, #feData.Axes[4] do
sum_Hphi=0+i0
Hphi=0+i0
-- Sum Hphi over phi
for v_count = 1, nphi do
Hphi = feData[freq_count][u_count][v_count][n_count].hfieldcomp2
sum_Hphi = sum_Hphi + Hphi
end -- v_count
-- Set the values in the new dataset
iSumData[freq_count][u_count][n_count].Ix = 2math.pirho*sum_Hphi/nphi
end -- n_count
end -- u_count
end -- freq_count

-- inspect(iSumData)
ISumStoredData = iSumData:StoreData("Custom")

ISumStoredData.Label = app.StoredData:UniqueName("Ix_" .. string.gsub(model, "%p", "_"))

All comments

Torben Voigt

Hello Tobias,

The Method of Moments (MoM) only calculates the currents on the boundary surfaces of objects. In the case of metallic objects, the currents only run on the surface anyway; in the case of dielectric objects, the Surface Equivalence Principle (SEP) is applied. In this case, the currents on the surfaces represent the currents that would actually run inside.

Accordingly, you can only calculate surface currents.

You might expect that FEM or MoM VEP (Volume Equivalence Principle) would also allow you to analyze volume currents, but unfortunately this is not supported. The exportes .ol and .os files will be empty.

You can request (surface and line) currents in CADFEKO here:

Best regards,
Torben

tobias_from_edag

Yes, I know that the currents are only nonzero on the surfaces. I was hoping that a post-processing macro was possible to find those surfaces that penetrate the target plane and calculate their sum. But as it seems impossible to do that using the documented post-processing api, I have given up.

I am currently attempting to write a field macro to calculate the currents using ampere's law after defining a near-field request on the target plane. I was hoping such a macro already exists in some form, like it does for other field simulation programs.

Is there some sort of recommended package for such vector math operations provided with feko? Or is it necessary to import an external lua package or to code the integration by hand?

Torben Voigt

Hi Tobias,
In this case I would recommend to create a support request on Altair One.

Best regards,
Torben

Georgi_GC

Hello, Tobias!

In POSTFEKO, in Results menu, submenu Requests, right click on CURRENTS and select "Add new" 3D View ( file).

3D.pdf

Set "Instantaneous magnitude" in Quantity "electric currents"'. To view arrows - select "Show arrows" in the "Result" tab.

tobias_from_edag

Hello future Readers,

I did end up re-asking this question to support and got an answer:

One can split the conductor and add an edge port to the split point. Then adding a voltage source allows one to read the current.

They also wrote a script that finds the current using Ampere's law from a previously defined nearfield :

-- Compute common mode current vs. distance along the cable
app = pf.GetApplication()

--Create a new (empty) result set to store commonm ode current
iSumData = pf.DataSet.New()

--Add quantities to the new result set
iSumData.Quantities:Add('Ix', 'complex', 'A') -- Ix

--Show the structure of the new dataset
print(iSumData)

-- Number of phi samples -- assume it is te same for all near-field requests
nphi = #feData.Axes[3]

-- Number of frequency samples -- assume it is te same for all near-field requests
nfreq = #feData.Axes[1]

-- inspect(iSumData)
ISumStoredData = iSumData:StoreData("Custom")

ISumStoredData.Label = app.StoredData:UniqueName("Ix_" .. string.gsub(model, "%p", "_"))