Learning Forum

[jure.soklic]

Hi everyone,

I have been playing around with matching circuits using lumped RLC boundaries recently, comparing the results with those of circuit simulators as well as analytical results. I designed a simple model of a 50 Ohm PEC sheet microstrip line on FR4_epoxy substrate with infinite ground, followed by a parallel RLC resonant structure (1st figure) and terminated with a PEC sheet connected to the ground plane. Using a single parallel RLC boundary, I could get results (2nd figure) which match well with the expected results (circuit simulation). However, when separating this boundary into three separate boundaries (3rd figure), as would be the case in practice, each representing one lumped component, I obtained completely detuned results (4th figure).

Unfortunately, I could not find any explanation for this behavior or the required modifications to my model when searching through software documentation and online. Can anyone give me an explanation as to why this happens and how an accurate simulation using separate RLC elements connected in parallel could be achieved?

Thank you in advance,

Jure

[Aymen Mzoughi]

It looks like BCs directly contact lumped port. Try to avoid this situation and allow the current to redistribute along trace before termination.