Modeling a Passive Sensor in HFSS

Tagged: ,

    • jac0119


      I am looking to model and simulate a passive sensor that is wireless. Though, I am not sure how to approach the port setup for this situation. The real-world sensor design has no connections and the desired way to operate it is to sweep it with a separate antenna to extract the passive sensor's resonant peak. The passive structure is composed of a square planar spiral inductor and a parallel plate capacitor on the same substrate (LC).

      To me, at least, a lumped port or wave port on the passive structure itself does not make sense because the real-world equivalent will not have physical connections and the results we have been getting are not as expected. I have also tried modeling a separate antenna in the solution space that would interact with the passive sensor, but the results here did not seem right either.

      If one of the methods of modeling mentioned would be the a more fitting method for this application, then it is highly likely that the port set up is not correct. Though, we are also not entirely sure what an accurate method of modeling would be for this scenario, as none of us on the project are well-versed with Full-wave FEM simulators.

      So, could we have some advice on how to approach this problem? I can attach HFSS archives and pictures if requested.

      Thank you.

    • Praneeth
      Ansys Employee

      You can use incident waves excitation type available in Ansys HFSS. For more details regarding this, please refer "Incident Waves" section in the HFSS help document.
      All the very best.

    • jac0119
      Thank you, The fields are looking really good.
      Do you know how you would obtain the S-parameters from this setup? I've looked in the help document and could not find anything nor any helpful online results.
    • jac0119
      I just wanted to give information on where I'm at after working on the model and reading online resources. I have the zero phase of the incident plane wave at the origin, which is also the bottom center of the sensor's substrate. The incident waves have a khat of <0,0,-1>. The polar plots from this look good.
      Currently, I am attempting to use the field calculator to calculate net power flow through non-model sheets (the process I am using is on page 16 of the resource I attached.)
      I want to use the calculated power so that I can define an output variable, S11, as 10*log(ReflectedPowerInput/IncidentPowerInput) and plot that over a frequency range.
      I am unsure how to use a non-model sheet for this, however, because I think that the net power through the sheet, in this case, will be affected by both the incident waves coming in and the scattered waves going out. So, I am unsure how to differentiate.
      An idea that I will try tomorrow is to use the separate sides of the sheet as the incident and reflected (top surface being the incident, and the bottom surface facing the sensor being the reflected.
      If anyone has any ideas, or seeing that I am approaching this problem incorrectly, please let me know. It would be greatly appreciated!
Viewing 3 reply threads
  • You must be logged in to reply to this topic.