# Properties of RF harmonics

• Plat

#### Plat

How can the power/amplitude of a particular RF harmonic be calculated? I would assume it is some well-defined fraction of the amplitude of the main frequency?

Do RF harmonics from a square-wave drive appear only on the even, odd, or both, multiples of the base frequency?

How do high-frequency harmonics propagate through electronic components? For example, if a MOSFET can switch at 1MHz max, then can it convey harmonics from the source/signal frequency that are above 1MHz? I know I'm greatly simplifying things here.

Fourier Analysis enables us to calculate the amplitude of each harmonic in a repetitive/periodic wave.

Do RF harmonics from a square-wave drive appear only on the even, odd, or both, multiples of the base frequency?
A square wave has an infinite number of odd harmonics. The n'th harmonic has an amplitude of 1/n compared with the fundamental.
https://en.wikipedia.org/wiki/Square_wave#Fourier_analysis

How can the power/amplitude of a particular RF harmonic be calculated? I would assume it is some well-defined fraction of the amplitude of the main frequency?
In general, RF harmonics are a bad thing, especially in communication systems. Quiz Question -- Why? And there is no general way to calculate them. They arise from several different mechanisms, some generating odd harmonics some generating even harmonics, and some generating both.

How do high-frequency harmonics propagate through electronic components?
A circuit has a transfer function that shows circuit gain and phase shift across frequency. In linear circuits, the fundamental and each harmonic must be treated separately as each passes through the circuit. The resultant signal will be the sum of all the differently shifted and scaled sinewave harmonics.
The spectrum of the output signal is the spectrum of the input signal multiplied by the transfer function.

For example, if a MOSFET can switch at 1MHz max, then can it convey harmonics from the source/signal frequency that are above 1MHz?
A MOSFET needs to be on or off most of the time to minimise both heating during each transition and the power needed to charge and discharge the gate. For that reason, switching rate is meaningless without qualification. It is unlikely that a switching MOSFET application would be driven by a sinewave alone.

How do high-frequency harmonics propagate through electronic components?
It would depend entirely on the 'frequency response' of the system you are dealing with. There is nothing magic about harmonics. They are totally independent of the fundamental, once they have been created. Their frequencies will stay the same - that's all.
To find out more about the relationship between the shape of a waveform in time and its description in 'frequency space', try this wiki article. Frequency and time domain descriptions are just alternative ways of describing the same signal. It is not 'really' one or the other.