The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> SP simulation and negative real Y
https://designers-guide.org/forum/YaBB.pl?num=1218635777

Message started by xristigi on Aug 13^th, 2008, 6:56am

Title: SP simulation and negative real Y
Post by xristigi on Aug 13^th, 2008, 6:56am

Hello everybody,

I have been running sp simulation to find the equivalent parallel impedance in a node of my circuits and it seems that :

Yreal=-1m
and Yimag=3m

Should I worry for this negative real part?? Does it imply instability??
What does it mean exactly?

Thanks in advance

Title: Re: SP simulation and negative real Y
Post by pancho_hideboo on Aug 13^th, 2008, 8:38am

xristigi wrote on Aug 13^th, 2008, 6:56am:

I have been running sp simulation to find the equivalent parallel impedance in a node of my circuits and it seems that :
Yreal=-1m
and Yimag=3m
Should I worry for this negative real part?? Does it imply instability??
What does it mean exactly?

It could be conditionally stable.

Title: Re: SP simulation and negative real Y
Post by xristigi on Aug 13^th, 2008, 9:02am

Thanks,

I also came to that conclusion... I also realized that it is due to the emitter followers that I have connected there. I guess I should check stability for the given source impedance.

Thanks again

Title: Re: SP simulation and negative real Y
Post by pancho_hideboo on Aug 13^th, 2008, 9:19am

xristigi wrote on Aug 13^th, 2008, 9:02am:

I guess I should check stability for the given source impedance.

You have to check stability both for source and load impedances.

Title: Re: SP simulation and negative real Y
Post by buddypoor on Aug 13^th, 2008, 10:12am

Quote pancho_hideboo: It's conditionally stable

Conditional stability is a specific term from the control theory describing a system which can be stabilized by increasing the loop gain.
Presently, I cannot see how one can come to the conclusion that a negative real part of the input conductance of a system indicates this kind of "conditional stability". Perhaps something else is ment ?
Question to pancho_hideboo: What does this mean for the system under discussion ? What has to be done to destabilize the system or to keep it stable ?

Title: Re: SP simulation and negative real Y
Post by pancho_hideboo on Aug 13^th, 2008, 5:04pm

buddypoor wrote on Aug 13^th, 2008, 10:12am:

Conditional stability is a specific term from the control theory describing a system which can be stabilized by increasing the loop gain.

No, you are wrong. Maybe you are not familiar with RF circuit theory.
Terms, "Conditionally Stable" and "Unconditionally Stable" are also used in RF circuit.
But these can not be completely related to stability theorem in control system theory.

http://www.designers-guide.org/Forum/YaBB.pl?num=1190272820

We can judge whether an amplifier is unconditionally stable or not by well-known K-factor or Mu-factor.

See stab_fact() and mu() in "Chapter 9: S-Parameter Analysis Functions" of the following.
http://cp.literature.agilent.com/litweb/pdf/ads2005a/pdf/expmeas.pdf

buddypoor wrote on Aug 13^th, 2008, 10:12am:

Presently, I cannot see how one can come to the conclusion that a negative real part of the input conductance of a system indicates this kind of "conditional stability". Perhaps something else is ment ?

Assume a simple amplifier. Here I define followings.
Gamma_S=Source Reflection Coefficient
Gamma_In=Input Reflection Coefficient
Gamma_L=Load Reflection Coefficient
Gamma_Out=Output Reflection Coefficient

StabFactor_S=real(Gamma_S*Gamma_In)
StabFactor_L=real(Gamma_L*Gamma_Out)

If both StabFactor_S<1.0 and StabFactor_L<1.0 are satisfied, this amplifier is stable. This is conditionally stability.
And this conditionally stability could be satisfied even if input impedance or ouput impedance have negative real part.
This stability is dependent on source and load impedances.
For example, when an input of amplifier is left open or short, abs(Gamma_S)=1.0. If this amplifier has real(Yin)<0 or real(Zin)<0, abs(Gamma_In)>1.0. So this amplifier could not be stable.

buddypoor wrote on Aug 13^th, 2008, 10:12am:

What does this mean for the system under discussion ? What has to be done to destabilize the system or to keep it stable ?

For example, an amplifier of real(Yin)<0 could be stable if shunt conductance is inserted at input of amplifier so that it cancels real(Yin)<0.
Of cource, this too simple stabilizer degrades gain of amplifier.