Why is Vth negative?

Hi there,

I'm a student working on power amplifiers designs who just started recently on master. I've had some background in analog IC in my undergraduate courses. but I have some question about the FETs used in RF PA that are puzzling to me

One of the question I have is why RF PA (such as GaAs or GaN) have negative Vth as stated in data sheet such as http://www.cree.com/Products/pdf/CGH40010-Rev1_5.pdf. This really does not make sense to me since we're dealing with NMOS are we not?

also, do these transistor exhibit early effects like silicon cmos?

RF textbook such as Cripps' seem to focus on the "knee voltage". Is this voltage the same as Vds(sat) in analog IC? (i.e the voltage under which transistor enter saturation from linear region)?

also, why is it that its not possible to observe RF voltage and current waveforms in real life. I'm told we can only measure power, why is that?

Lastly, what is the difference between linear gain and power gain? Arent we always talking about power when we say dB or dBm? (i.e 20log(voltage) or 10log(power)). why make the distinction?

I realize i'm asking a lot of questions. but theses fundamentals are really bothering me

Thanks in advance to anyone who can answer these question

Just an obvious point. The said transistors, n-channel, have actually a negative Vth cause they are depletion type.

I assume, you are talking about GHz range. Voltage measurements should be possible up to several GHz bandwith, but not without causing a considerabe load impedance to the test object.

Hi,

Thank you for letting me know about depletion MOS. I've only dealt with enhancement MOS before.

At low frequency we're taught that with high impedance probe we are able to observe voltage waveform without loading the circuit.

I'm speculating that high impedance probe will cause reflection at high frequency hence it cannot be used?

Suppose we use 50 Ohm terminated probes at high frequency, but such would load the circuit heavily. But wouldnt we be able to correct for the loading effect hence get an accurate measurement anyways? Is it because it is very difficult to properly manufacture purely resistive probes?

If my understanding is correct, many amplifier classes are defined by their voltage and current waveforms and respective harmonic termination.
It seems puzzling to me that we're not able to measure voltage and current waveform at high frequency(hence verify that, say, a class F is actually behaving as class F)

Are there more fundamental reasons why people havent been pushing to measure V/I at high freq, and simply rely on power measurements ?

Thanks