Class E RF PA questions

In order for a Class E RF PA to have a linear or even near linear output, must it always have an envelope EER circuit driving the supply voltage also?

I mean, if I don't do that then I'm basically broadcasting only the phase changes in the signal and not the amplitude changes, right?

If I try to drive the MOSFET with the lower amplitude portions of the modulated RF input then wouldn't the MOSFET conduct less, shut OFF sooner and then when the capacitance across the Drain finishes discharging, the output waveform would swing negative before the MOSFET gets turned ON again on the next cycle?

Do some people do it like that?

I've heard you can detect SSB from phase changes only so I'm wondering if people are using Class E RF PAs without any supply voltage modulation and it still works anyway because they are broadcasting mainly the phase changes?

Or do most people modulate the supply voltage?

I see a lot of information about modulating the supply voltage for AM transmissions but not for SSB.

Why is that?

Thanks if someone can enlighten me, George

You will get some response to AM modulation on the gate drive, but it will be highly nonlinear and the efficiency of the amplifier will drop as the input power drops.

AM when correctly modulated does not cross the zero power line (At most the envelope just touches it), and designing a drain modulator for this condition is very much easier then designing one for modes like SSB where the envelope does cross the zero power line.

A Class E stage where the mosfet is not swiching at zero voltage/zero current is not a class E stage.....

I am working on a (Non class E) two quadrant polyphase converter for an experiment in high(er) efficiency HF radio, the output stage however is broadband and so class E is out.

Regards, Dan.

If I'm doing PSK31 on SSB, or even phone(voice) on SSB, can it still work to have a fixed Drain supply voltage, even though the output is non-linear?

I mean, a lot of people push their voice modulation way up to get their signal out and sound louder anyway, right.

If the digital mode is bi and not quad or higher, it ought to still work, right?

It is only digital modes that use multiple amplitude levels that would get distorted, right?

I mean, as long as the signal is either plus or minus but of same amplitude, it ought to work, right?

- - - Updated - - -

I guess I've had this thread in the back of my mind while thinking about all this:

http://forums.qrz.com/archive/index.php/t-405552.html

I remember reading that a long time ago.

- - - Updated - - -

Dan Mills said:

I am working on a (Non class E) two quadrant polyphase converter for an experiment in high(er) efficiency HF radio, the output stage however is broadband and so class E is out.

Regards, Dan.

Click to expand...

I hope your experiment is sucessful. I looked up polyphase converters and I'm not catching on right away.

Sorry if all my questioning comes across the wrong way. I'm still trying to get my head wrapped around all this as I haven't designed my own RF PA before so now I'm finally thinking about it all in more depth.