Class C amplifier at 5 GHz

[OZZAA]

I tried to start designing a Class C power amplifier at 5 GHz (for any vlaues of gain and output power)
I used TSMC 180 nm technology.
I build the following circuit to test some concepts

I found that the max available gain is in -dB (Attenuator)
also when I modified the terminals' impedances as shown in the circuit the S21 is also -dB (attenuator)

the gate bias voltage is 0.1< (Vt=0.5 V) as shown in the following figure


What is the problem
If the gain is still -ve the PAE will go to -ve values, is not it?

Any Help will be highly appreciated

 

[BigBoss]

Small Signal Analysis of a C-Class Amplifier doe not make sense because drain current does not start to flow till just before input signal reaches a level so that bias condition appears.
Instead, Define your current flow angle,select a bias voltage and consequently a drain current the make a HB analysis under large signal condition then compare the powers ( input and output) finally find the Gain.
Designing C-Class Amplifier is really tough task and if you don't have strong experiences over Power Amplifiers, it will take so long time to optimize the amplifier.

 

[mtwieg]

Yeah S parameters are basically meaningless for a class C amp. You need to do load and source pull simulations with PSS/HB.

 

[OZZAA]

Instead, Define your current flow angle,select a bias voltage and consequently a drain current the make a HB analysis under large signal condition then compare the powers ( input and output) finally find the Gain.
Designing C-Class Amplifier is really tough task and if you don't have strong experiences over Power Amplifiers, it will take so long time to optimize the amplifier

Yeah S parameters are basically meaningless for a class C amp. You need to do load and source pull simulations with PSS/HB.

I followed your appreciated previous advice.
I designed this power amplifier, however the gain is not constant as shown in the attached figure.

I choosed the Bias gate voltage = 0.3 V, while the threshold voltage approximately 0.5 V.
why the power gain is not constant before the compression point?

 

[mtwieg]

That looks like fairly reasonable behavior for a class C (hard to tell without more context)
For a class C amp, it takes some minimum amount of input power to cause the transistor to do anything. Below that power level, the conduction angle is zero, and gain is also
As you increase input power, the conduction angle will increase, and the gain will increase along with it.
At some point, if you raise input power enough, the transistor will be basically driven as a switch. Beyond that point, you can't get any more output power out of it. So gain must decrease (probably long before you get to that point).

If you want an amplifier that has flat gain vs power level, then you don't want class C (unless you're willing to put effort into predistortion).