Class E and Class F Amplifier

[AMSA84]

Hello guys,

Can someone explain how class E and class F amplifiers work? I tried to look around the web, but I didn't understood.

For example, in class F, the square waveform that we can obtain using the filter to get the fundamental (in parallel with the mosfet at the output) and a series LC tank circuit to get the 3rd harmonic, appears at the output? Or at the output appears only a sinusoidal waveform?

 

[KlausST]

Hi,

it is very difficult to find... I´m sure you spent weeks of seeking that information....

https://en.wikipedia.org/wiki/Amplifier#Additional_classes

Klaus

 

[vfone]

https://www.youtube.com/watch?v=iABwHeZ3_Jw
https://www.youtube.com/watch?v=Ps8W6vMbItM

 

[AMSA84]

Hi, thank you for the videos. Very helpful.

Allow me to ask you one question. In the Class-E topology, those almost square waveform of the current and voltage (non-overlapped) are in the transistor. They don't show the output waveform. How's like? Is it sinusoidal? I was interested in using those two non overlapped signals (current and voltage) from this topology. That said, if the output is sinusoidal, and I don't want that, how can I take the advantage of those two signals ZVS in the transistor?

 

[vfone]

In Class-E the transistor behaves as a perfect switch. When it's ON, the collector/drain voltage is zero, and when it's OFF the collector/drain current is zero. So, the shape of the waveforms cannot be sinusoidal.

 

[AMSA84]

Hi, thanks for the reply. vfone, but that behavior is on the MOSFET (in the switch). What about the output voltage? How's the output waveform shape at the load? It is not sinusoidal? Because of the resonant LC circuit?

 

[mtwieg]

The canonical class D, E, and F topologies (and their inverses) are all designed to produce power at their fundamental frequency, with a sine output. The actual shape of the drain current/voltage waveforms will be very different, in order to get high efficiency. By putting filters on the output, nearly all the output power can be constrained to the fundamental.

If one wanted, I suppose it may be possible to get non-sinusoidal waveforms at the output, though I presume doing so would require extremely sophisticated filters. In particular if someone were to devise a class E amp with a square wave output, I think that would be incredibly useful.

 

[KlausST]

Hi,

The canonical class D, E, and F topologies (and their inverses) are all designed to produce power at their fundamental frequency, with a sine output.

Afaik class D with a filter is made to produce DC at its output (with a constant PWM). By varying the pwm duty cycly it is possible to vary the output voltage. With this you may produce about any output waveform Therefore class D is often used as an audio amplifier.

Klaus

 

[mtwieg]

Hi,



Afaik class D with a filter is made to produce DC at its output (with a constant PWM). By varying the pwm duty cycly it is possible to vary the output voltage. With this you may produce about any output waveform Therefore class D is often used as an audio amplifier.

Klaus

Class D refers to two different things depending on context. In RF/microwave it's a carrier amplifier with two switches always operating near 50% duty cycle. In a more general context (like for audio and power applications) it's what you describe. In RF/microwave, this is usually referred to as a class S amp.

 

[AMSA84]

Hi guys, thank you for replying to my question.

mtweig, the idea was to use a class E or F topology to drive a big transistor, as for example a power transistor.

However, I was wondering how that would be possible if we think that at the output of those types of amplifiers the output is a sinusoidal waveform. The main focus has to do with the attempt to get an ideal switch (as far as possible, of course, since it is a resonant topology), when there is current, there is no voltage and vice-versa.

So I don't know if this might be a good solution. It is possible to have a squarewave at the output of this types of amplifiers?