Does a filter or inductor's power rating matter in its passband, or only in the attenuated frequency range?

Question

I would like to build a high-pass filter that would block reception of signals from the United States' FM broadcast range (88–108 MHz) while still allowing me to transmit VHF ham signals (around 146 MHz). Using https://rf-tools.com/lc-filter/ generates a circuit like this:

The way I understand this circuit is that the signals above the cutoff go through the capacitors, while the signals below the cutoff go through the inductors. So if I intend to transmit through this circuit I imagine I would need capacitors rated to handle say a peak-to-peak range of 75V or so.

But what about the inductors? On the one hand, there shouldn't be much current flowing through them at all. But on the other, they would still be exposed to the overall voltages. It appears that inductors are rated primarily by current — is it safe to assume that the risk of arcing or whatnot is negligible, and that my intuition about not needing to handle the passband's transmit current is correct?

In a similar vein, there are commercial FM bandstop filters but they are not rated for anywhere near my required 5–10W transmit power. But their insertion loss at my transmit frequency is minimal. If the amount of power dissipated through loss in the passband is less than the rated power handling, would an "RX" bandstop filter be safe to use for transmission as well?

Answer 1

Signals very far above the passband indeed only "pass through" the capacitors, and only the capacitors' current rating and the inductors' voltage handling matter in that case. However, pay attention to the self resonant frequency (SRF) of the inductor, above which they start to act like lossy capacitors.

Closer to the passband but still above it (think 50% above the 3dB frequency), the filter behaves more like a cascade of impedance matching networks, with each L-C leg transforming to another impedance, then the last L-C leg transforming it back to 50 ohms. The voltage across an inductor can be higher than the input voltage in this case, but not by much.

I recommend simulating the circuit in LTSpice to see the voltages and currents across the inductors. If you've measured the insertion loss at the frequency of operation and deemed the power dissipation acceptable, then the only risk is arcing over, which is unlikely below 10W.