simulation problem with coupling between two cables

Sam_21831
Sam_21831 Altair Community Member
edited March 22 in Community Q&A

Hello,

I am testing the coupling between two cables but the simulation results show a significant differences with the measurement data.

Please find model and measurement results attached.

 

 

 

 

Tagged:

Answers

  • Torben Voigt
    Torben Voigt Altair Community Member
    edited March 15

    Hi Sam,

    I believe that it is recommended to use only one harness and only one cable path in case of strong coupling of neighboring (and parallel) cables. Could you try that?

    Best regards,
    Torben

  • Sam_21831
    Sam_21831 Altair Community Member
    edited March 15
    Torben Voigt_20420 said:

    Hi Sam,

    I believe that it is recommended to use only one harness and only one cable path in case of strong coupling of neighboring (and parallel) cables. Could you try that?

    Best regards,
    Torben

    Hi Torben,

     

    Could you explain this better?

     

    Sam

  • Torben Voigt
    Torben Voigt Altair Community Member
    edited March 18
    Sam_21831 said:

    Hi Torben,

     

    Could you explain this better?

     

    Sam

    Hi Sam,

    I was thinking of having only on cable path, one cable bundle (source and victim cable) and one harness. The distance between the cables can be defined in the bundle:

    image

    The harness would look like this:

    image

    I attached the edited model (Feko version 2023.1.1).

    Best regards,
    Torben

  • Sam_21831
    Sam_21831 Altair Community Member
    edited March 18
    Torben Voigt_20420 said:

    Hi Sam,

    I was thinking of having only on cable path, one cable bundle (source and victim cable) and one harness. The distance between the cables can be defined in the bundle:

    image

    The harness would look like this:

    image

    I attached the edited model (Feko version 2023.1.1).

    Best regards,
    Torben

    Hi Torben, 

     

    Thank you for help and support. I used what you have recommended but still realize big differences between Feko simulation results and the measurement data.

    For example, this is the comparison between SIM and EXP for 0.5 in. spacing

    image

     

    Regards,

    Sam

  • Sam_21831
    Sam_21831 Altair Community Member
    edited March 18
    Sam_21831 said:

    Hi Torben, 

     

    Thank you for help and support. I used what you have recommended but still realize big differences between Feko simulation results and the measurement data.

    For example, this is the comparison between SIM and EXP for 0.5 in. spacing

    image

     

    Regards,

    Sam

    Hi Torben, 

     

    One more thing, I have tried to test the unshielded cables. I realized at 1MHz (as an example) that the calculated voltage across the victim resistor (in mV) is about 25.97 but for the previous test (one cable is shielded) the voltage is 46.29 mV. This is unexpected as this value should be less. 

     

                         image

     

     

     

    Thanks,

    Sam

  • Torben Voigt
    Torben Voigt Altair Community Member
    edited March 19
    Sam_21831 said:

    Hi Torben, 

     

    One more thing, I have tried to test the unshielded cables. I realized at 1MHz (as an example) that the calculated voltage across the victim resistor (in mV) is about 25.97 but for the previous test (one cable is shielded) the voltage is 46.29 mV. This is unexpected as this value should be less. 

     

                         image

     

     

     

    Thanks,

    Sam

    Hi Sam,

    The model I attached yesterday (shielded_one_harness_one_cable_path.cfx) is based on your model in the shielded_0p02 folder. The distance between the cables is 0.0015424 m = 0.0607 inch. I don't find a related result in your .xlsx file.

    To reduce complexity, could you please attach one model here with related measurement data? 

    I quickly tested shielded vs. unshielded and the voltage is as expected:
    image

    Best regards,
    Torben

  • Sam_21831
    Sam_21831 Altair Community Member
    edited March 19
    Torben Voigt_20420 said:

    Hi Sam,

    The model I attached yesterday (shielded_one_harness_one_cable_path.cfx) is based on your model in the shielded_0p02 folder. The distance between the cables is 0.0015424 m = 0.0607 inch. I don't find a related result in your .xlsx file.

    To reduce complexity, could you please attach one model here with related measurement data? 

    I quickly tested shielded vs. unshielded and the voltage is as expected:
    image

    Best regards,
    Torben

    Hi Torben,

     

    Thank you again for your help. Here are attached the case of shielded and unshielded cables I was testing on Feko with a separation distance of 0.5 in. (please note that this distance is taken from the outer radius of the cable plus the coat thickness). I also, provided an excel sheet for the corresponding measurement data where I don't see an agreement between both resulted numbers.

     

    Thank you

     

    Regards,

    Sam

  • Torben Voigt
    Torben Voigt Altair Community Member
    edited March 20
    Sam_21831 said:

    Hi Torben,

     

    Thank you again for your help. Here are attached the case of shielded and unshielded cables I was testing on Feko with a separation distance of 0.5 in. (please note that this distance is taken from the outer radius of the cable plus the coat thickness). I also, provided an excel sheet for the corresponding measurement data where I don't see an agreement between both resulted numbers.

     

    Thank you

     

    Regards,

    Sam

    Hi Sam,

    I get quite good agreement when dividing your measured voltage in the victim cable by 2 (your values are peak-to-peak, Feko uses peak values!) while still keeping 3 V as source:

    image

    This is a bit surprising indeed, since I would think that I'd have to use 1.5 V as source. 

    I will investigate further.

    Best regards,
    Torben

  • Torben Voigt
    Torben Voigt Altair Community Member
    edited March 22
    Torben Voigt_20420 said:

    Hi Sam,

    I get quite good agreement when dividing your measured voltage in the victim cable by 2 (your values are peak-to-peak, Feko uses peak values!) while still keeping 3 V as source:

    image

    This is a bit surprising indeed, since I would think that I'd have to use 1.5 V as source. 

    I will investigate further.

    Best regards,
    Torben

    Hi Sam,

    Since the problem you are having is also being dealt with in a support case by my colleague Shannon Mistry, I would suggest that we put this on hold. If you have a solution from Shannon, perhaps you could post it here so that the community can benefit from it?

Categories