Signal Amplitude problem due to TVS diodes, Impedance matching problem

RE: Signal Amplitude problem due to TVS diodes, Impedance matching problem

aswinvenu — Sun, 28 Jul 2024 19:09:36 GMT

Just out of curiosity, Why do you want to match the impedance? Is 5MHz the maximum frequency your circuit going to deal with?
Can you share the part number for 2KOhm ferrite bead? 2KOhm impedance is usually rated at 100MHz. If your signal is maxed at 5MHz the ferrite bead's impedance will be quite low. If you really want to match the resistance, better of using a resistance in series! ( I don't know why it's required )

Like [mention:4f57fc9d538949ad9eb336ddb9469bb8:e9ed411860ed4f2ba0265705b8793d05] has mentioned the voltage drop is probably due to the junction capacitance. At high frequencies it provides a low impedance path to ground.

RE: Signal Amplitude problem due to TVS diodes, Impedance matching problem

jc2048 — Sun, 28 Jul 2024 08:43:01 GMT

Reverse biased, the TVS is a capacitor (thin insulating layer - the junction - with charge either side). This is what ST give in a datasheet for a 400W SMA part

So you're loading your output with anything from a few hundred pF to 1.5nF, depending on the voltage of the part.

You need to look for a part designed for protecting signal lines (generally, they'll incorporate a second low-capacitance signal diode that 'shields' the capacitance of the TVS part until it avalanches, or sit the TVS in a bridge for the bidirectional one). Perhaps look at what's available for protecting RS485 lines, or possibly USB if the 5V isn't too restrictive for you.