Measuring gain of an RF amplifier?

[elroy]

I have an RF amplifier that's designed to be used between an antenna and the input to an HF receiver or SDR. I'd like to measure its gain at various frequencies. I have a waveform generator (Siglent SDG2042X) and an oscilloscope (Siglent SDS1202X-E).

So I'm thinking I'll set up the generator to put out a signal at a given frequency and Vpp (picking a level similar to that of a signal in from an antenna), feed it into the oscilloscope (using a terminator for impedance matching purposes) and see what the scope reports as the voltage of the waveform, then insert the amp between the generator and scope and measure again, then convert that into dB.

Does that seem like the best way to accomplish this? I also have a miniVNA, but unclear to me if it would be useful in this case.

[kg4arn]

Sounds to me like that should work but you have to keep the output of your rf amp well below the 1 dB compression point. (P1dB - at least 10 dB).  Is this a homemade amp? 

[vk6zgo]

Nothing wrong with "old school".
It is probably best to check the gain at a number of "spot frequencies" over the range you are interested in, though.

[xrunner]

Here's some info on the accepted "standards" for S-Meter readings for HF. Might give you some idea of the dBm signal levels you'd be receiving in a real world situation.

S Meter Levels

[pigrew]

Normally, I'd suggest a VNA to do these measurements since it'll tell you the mismatch of the amplifier (S11 and S22), but it maybe difficult with the MiniVNA as it looks like it has fixed output power, and in your case that power level could overpower your LNA (does it have a maximum input power specified?).

You will need to add an attenuator to reduce the power level at the input of your LNA so that you're working at the proper power level. S21 is the gain of the amplifer, exactly the figure you want. With the attenuator, your S11 measurement accuracy will be greatly degraded. Note that S21 is the voltage gain, assuming 50-ohm ports on the input and output. If your antenna or receiver are not exactly 50 ohms, the gain will be somewhat different in its final application.

If you go the oscilloscope route, I suggest having it measure "V_RMS cycle-to-cycle" when in AC coupling mode, and not Vpp. Vpp is much more influenced by noise in the oscilloscope. Also note that the siggen's output will be different amplitudes at different frequencies (especially towards its high end). You'll want to measure your THRU standard's V_rms at every frequency you'll measure. The oscilloscope's ADC is 8-bit, so you'll need to switch ranges to get a good measurement with and without the amplifier.  Switching ranges will add a percent or two of inaccuracy in the gain. With the VNA, you would not need to change ranges, allowing you to make a more accurate gain measurement. At your frequencies, an external feedthru termination is fine, and the input capacitance of the oscilloscope is negligible.

As KG4ARN said, keep way below your P1dB. Going towards it will hurt your LNA's linearity (among other things).

PS: With fancier VNA's, testset ports are broken out so that the attenuator can be added on the VNA side of the coupler, allowing for better measurement accuracy. The couplers normally can handle more power than the receivers.

[elroy]

Thanks for the suggestions. I liked the idea of working with a scope because it's simple, but I can see there could be benefit to using the miniVNA. Maybe I can try both as a learning experience and compare the results.

The amp is neither homebrew by me, nor a regular commercial product. It was designed and fabricated by an EE who does them in small lots and sells to customers on demand.

I also have a 110 dB variable attenuator, and thought I'd use that to ensure that the signal levels are as low as they need to be. I've noticed that even at its minimum the SDG2042X's output is on the hot side to feed to something like an HF receiver's antenna input.

[tautech]

Thanks for the suggestions. I liked the idea of working with a scope because it's simple, but I can see there could be benefit to using the miniVNA. Maybe I can try both as a learning experience and compare the results.

The amp is neither homebrew by me, nor a regular commercial product. It was designed and fabricated by an EE who does them in small lots and sells to customers on demand.

I also have a 110 dB variable attenuator, and thought I'd use that to ensure that the signal levels are as low as they need to be. I've noticed that even at its minimum the SDG2042X's output is on the hot side to feed to something like an HF receiver's antenna input.

Still, you can pad down the 2042X with that 110dB attenuator and better match what the RF amp expects to see.
IIRC SDG min output is 2mV into 50 ohms.
Have a look at TSP videos for doing things like this as he often does these sort of exercises when reviewing gear.

[PA0PBZ]

I also have a 110 dB variable attenuator, and thought I'd use that to ensure that the signal levels are as low as they need to be.

So after inserting the amplifier simply adjust the attenuator to show the same amplitude on the scope and you have a direct readout of what the amplifier does