[SOLVED] how to determine RF receiver input impedance

[jdbrinton]

Hello,

I'm making a very low bandwidth 40MHz receiver. The receiver is fairly simple: an LC input filter and two FET gain stages go into an ADC. I'm using an "electrically small" strip-line antenna which of course has a tiny input impedance: ~1.96 -j1758. Do I need to add a matching network even though it's at such low frequency? The antenna is just centimeters away from the receiver so I'm pretty sure that the PCB trace impedances don't matter. How do I determine the input impedance of the receiver?

Thanks!!!
joel

 

[radiohead]

Use a VNA to measure the input impedance and design a matching network. Or at that frequency you can rely on the fet device model from the manufacturer.

 

[jdbrinton]

thx, radiohead! I got the model from the manufacturer but at that frequency input impedance is ~1.8kOhm. That makes the matching network inductor pretty huge (1,800 : 2), just wanted to do a sanity check to see if this seems right.

using the Phillips BF1210

 

[FvM]

You don't need to treat a few centimeters as transmission line, but the capacitance can matter though, because it's in the antenna impedance range.

You'll most likely want an inductor to compensate the antenna capacitance. A receiver amplifier doesn't necessary need impedance matching, because noise minimization is the primary objective. If a wide bandwidth requirement adds, a non-resonant active antenna design is suggested. But for small bandwidth, simple LC network impedance matching is the most preferred way.

I agree about extracting the amplifier input impedance from known transistor parameters.

 

[radiohead]

You don't need to treat a few centimeters as transmission line, but the capacitance can matter though, because it's in the antenna impedance range.

You'll most likely want an inductor to compensate the antenna capacitance. A receiver amplifier doesn't necessary need impedance matching, because noise minimization is the primary objective. If a wide bandwidth requirement adds, a non-resonant active antenna design is suggested. But for small bandwidth, simple LC network impedance matching is the most preferred way.

I agree about extracting the amplifier input impedance from known transistor parameters.

A few words for sake of clarity: One additional remark is that resonant antennas always have better efficiency (lower loss). But at 40 MHz it will be challenging to make a resonant antenna (free space wavelength of 7.5m). If the antenna is smaller the antenna will behave like a capacitor with a resistor in parallel. The better the efficiency, the higher this resistance. You'll have to match to this impedance.

 

[FvM]

If the antenna is smaller the antenna will behave like a capacitor with a resistor in parallel. The better the efficiency, the higher this resistance.

The original post gives a series equivalent circuit, which is the usual way to specify the radiation resistance. In a parallel equivalent circuit, an electrical short antenna appears as a capacitance with very high (100k range in the present case) parallel resistance.

The radiation resistance of 2 ohm would refer to a rod of about 0.5 m length according to RF handbooks. It's not clear if the 2 ohm value is a pure radiation resistance or contains losses.