How to measure a fast-varying RF power in a circuit?

Question

I have a circuit which generates an RF signal to drive a device, and I want to test it. For my application, I require the RF power to vary with a high frequency (of around 100 Hz), and the RF signal itself is in the range of 50-200 MHz.

I have a spectrum analyzer, but due to the fast varying of the magnitude of the RF power, I don't know whether I can observe the 'power envelope', ie., measure how the power varies, with a frequency of 100 Hz.

Is there any way in which I can reliably measure the variation of RF power envelope with time (100 Hz in this case)? What device is most suited for this purpose?

Answer 1

If you can set the RF signal to a fixed frequency, for example 200 MHz then set the SA's center frequency to that frequency (200 MHz), then use the "zero span" mode.

This will plot power over time (at the 200 MHz we set earlier) instead of the "normal" power over frequency.

Then you can adjust the sweep time so include a couple of 100 Hz periods, 50 ms would be a good choice.

It can be that you have to adjust the Resolution Bandwith (RBW) (and maybe Video bandwidth) to be able to select that 50 ms. Start with high RBW values (a couple of MHz).

If you cannot fix the RF frequency then the signal could be outside the SA's "window". At Fcenter = 200 MHz and RBW = 10 MHz that "window" would be 195 - 205 MHz. Most Spectrum Analyzers unfortunately do not have a higher RBW.

Answer 2

100Hz is NOT fast modulation.

Use an oscilloscope, possibly with a log amp on the front, but just a scope will give you the envelope in classic AM modulation monitor style.

The log amp will give you a log display which is far more useful for power measurement over a wide range (Typically you get 60dB or more of dynamic range that way).

If you use a digital scope, make sure the sample rate is sufficient to capture the RF peaks even with the slow timebase.

Answer 3

Use wide IF , zero LO sweep then show as time base scope with 10ms/div

Answer 4

The easiest way to the get a oscilloscope/digitizer with a suitable memory depth (assuming your scenario time is 1s or less).

You perform digitization on the signal, get the IQ data, then do a fourier Transform to get the PSD.

From there, you can use the data to calculate the power (depends on your channel definition)