[HFSS V14.0] How to match lumped ports?

Hello everyone. As a small test of the effect of port shape and size on their value, I've built a simple current loop of random dimensions with a capacitive element to make it a resonant structure at 128MHz. Looking at the eigenmode analysis result it behaves like it should. However, when I want to inspect its frequency behavior through a lumped port, the port is so mismatched that I don't get any result. I have to run the simulation, extract the device impedance, set the lumped port with the conjugate of this impedance, switch off port re-normalization and rerun the simulation. The resulting S11 graph shows the expected dip at the resonance frequency but also other odd behavior, like a >0dB baseline.

Is there some way in HFSS to have lumped ports automatically match the measured impedance? And does anyone have an idea of what is going on on this S-curve?