Microwave amplifier stability network

[jousse]

Hello

I am trying to design a MMIC amplifier at 30 GHz. When simulating K-factor (using StabFact knob) in ADS it is below 1 up to 35 GHz so I am trying to use a parallel RC network at the input and R at the gate bias to improve it. The problem comes when I finally achieve K>1 for all frequencies and I do LoadPull to the whole thing. The Gain falls from almost 12 dB (results with no RC and Rgate networks) to 5-6 dB which seems quite a big degradation.

Do you know if there is any way to know in advance if the stability networks are going to affect too much to the Load Pull results of Gain/Pout/PAE?

Many thanks in advance

BR
Jose

 

[vfone]

It is a challenge to stabilize a 30GHz amplifier using an RC network. Placing the RC network at the input of the amplifier you got about 6dB less gain, but for sure the noise figure of the MMIC increase to almost the same amount in dB. Try to place the RC network somewhere at the output of the amplifier. In this case the gain and the noise figure will be less affected, but linearity will decrease a bit.

 

[BigBoss]

Stability can also be improved by applying Negative Feedback.It does also not increase the Noise and gives a Stability and Gain Flatness simultaneously.If you apply both method ( RC and FB) you'll get flexibitly to play around Optimum Response.

 

[jousse]

Many thanks! Is there a way to quickly see how it affects to the response? Or do I need to get and simulate Load Pull each modification of the RC or feedback networks?