The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Aug 18th, 2024, 4:24am
Pages: 1
Send Topic Print
Small Signal Design (Read 3248 times)
nickc
New Member
*
Offline



Posts: 2

Small Signal Design
Jul 17th, 2006, 2:05pm
 
Hey

This forum seems to be for advanced designer but I have some basic questions.  I need to design a basic small signal amplifier for gain, not LNA or something else...just plain simple gain.  I have done the DC analysis (biasing), gain vs. frequency MSG/MAG analysis, and I can easily do matching networks (to 50 Ohms input and output).  What I am having trouble with is how to calculate input impedance and output impedance of the transistor (or transistor cells) so that I can match both to 50 Ohms for maximum power gain.  Please answer these questions below.  Almost of all you can easily answer these in a few minutes, and this will help me out much.  Please, if possible, answer each question so as to make it clear for me.

* First of all, does input or output impedance value depend on whether the other port is open or short or connected to a load?  I would think not, but I am not sure because of some odd results.

* Does Z11 equal input impedance? If that is true then isn't Z11 calculated with output open (i2=0), but then in actual amplifier output won't be open so how come Z11=input impedance (if it does).  This question also relates to the above question.

* The easiest way to calculate input impedance is to do V/I so I placed an AC voltage source (magnitude 1) and read the value of current and took the ratio Zin=1/I to calculate the impedance and I think this value is the correct one.  I calculated this with output not open; rather connected to a load (any reasonable size load, 50, 100 or 1000 ohm) or completely shorted.  Is this the right way to calculate input impedance?  This Zin that I calculated matched Z11 if the output, when calculating Z11, was shorted (which is not the definition of Z11).  Also, when I did Zin=V/I with output open and Z11 with output open (which is the actual definition of Z11, i2=0) then these two values matched.  If I don't want to do Zin=V/I then what is the right way and setup of measuring Zin using the parameters?  As far as I understand, Z11 with i2=0 (open output) does not give you Zin, correct?

* Does S11 or S22 value depend on what source is presented at the input or output? Why?  In other words, for example, if I do S22 without matched input and then do S22 with input matched, would these two values of S22 match each other or not?  

* Finally, the last question is about LNA.  I came across this while looking for answers.  In LNA, the input match is not conjugate matched but a source impedance is presented for optimum noise figure.  I understand that but then I read that the output is matched based on what source impedance is presented.  Why is that, would you just conjugately matched the output impedance to the 50 ohm load to transfer maximum power from outpu to the load?  What does that have to do with what source is presented at the input.  I know this question might be related to the one above it and answering that one might help me understand this one as well.

I am sure all these question arise from one or two basic underlying concepts.  Thank you very much for all your help...Nick
Back to top
 
 
View Profile   IP Logged
ACWWong
Community Fellow
*****
Offline



Posts: 539
Oxford, UK
Re: Small Signal Design
Reply #1 - Jul 18th, 2006, 2:35am
 
I think a quick answer to your query is that you are considering the amplifier as "unilateral" when in fact all real amplifiers are "bilateral", that is there is a finite s12 or z12 etc.

So input impedance Z11 or S11 does change depending on the load if Z12 or S12 is finite.

So to get the input impedance to match to (ie s11*) you should do it under the conditions of the correct loading. Measureing ac current in ac sweep to find input impedance is fine.

Sorry i'm a bit pressed for time now so can't come up with fuller explanation at the moment..... but try reading a book like Chris Bowick "RF Design"..

cheers

aw
Back to top
 
 
View Profile   IP Logged
nickc
New Member
*
Offline



Posts: 2

Re: Small Signal Design
Reply #2 - Jul 18th, 2006, 6:34am
 
Hey,

Thanks for th reply.  I'd still like somone to answer the first post but to add to your comments:

* I thought in small signal the input and output are not really correlated.

* But, in most cases amplifiers are bilateral so each terminal impedance affects the other--I understand that.  You mentioned the ac sweep (taking the V/I ratio) be done with appropriate load impedance.  In a 50 Ohm system what is the appropriate load impedance?  Is it 50 Ohm or is it output matching network connected to 50 Ohm?  As you can notice, if matching network is required then one would be running in circles because to find input impedance you need output match and to find output impedance you need input match.

* Also, as I have asked in the first post what is another appropriate way of finding impedance using the z/y/s parameters besides doing ac sweep?  Thank you

Nick
Back to top
 
 
View Profile   IP Logged
ACWWong
Community Fellow
*****
Offline



Posts: 539
Oxford, UK
Re: Small Signal Design
Reply #3 - Jul 19th, 2006, 4:52am
 
nickc wrote on Jul 18th, 2006, 6:34am:
Hey,

Thanks for th reply.  I'd still like somone to answer the first post but to add to your comments:

* I thought in small signal the input and output are not really correlated.

* But, in most cases amplifiers are bilateral so each terminal impedance affects the other--I understand that.  You mentioned the ac sweep (taking the V/I ratio) be done with appropriate load impedance.  In a 50 Ohm system what is the appropriate load impedance?  Is it 50 Ohm or is it output matching network connected to 50 Ohm?  As you can notice, if matching network is required then one would be running in circles because to find input impedance you need output match and to find output impedance you need input match.

* Also, as I have asked in the first post what is another appropriate way of finding impedance using the z/y/s parameters besides doing ac sweep?  Thank you

Nick


Hi Nick,

* i don't know what you mean by input and output being correlated ?? the input and output are definately interlinked, that is the whole purpose of two-port anaylsis that there exists S12 & S21 in addition to S11 and S22. (similar for Z, H, or any other two port parameters out there..)

* usually s12 is small and s21 is large. So although changing the output match will change in input impedance, the change should be small and tolerable. If changes drastically with small tweaks to the output match, you have a very bad amplifer. But envitablly one can often "go round the loop" a few times... to get an optimum solution.
So one normally matches the port most important first before attending to the other port. So for a power amplifier, one could match the output conjugately to the load (50 ohm or otherwise), whilst the input is say being driven from an appropriate source (say 50). Then having got the output match to the load right, one could go back to the input and match that up.

* i use sp analysis (from cadence spectre).

cheers
aw
Back to top
 
 
View Profile   IP Logged
Pages: 1
Send Topic Print
Copyright 2002-2024 Designer’s Guide Consulting, Inc. Designer’s Guide® is a registered trademark of Designer’s Guide Consulting, Inc. All rights reserved. Send comments or questions to editor@designers-guide.org. Consider submitting a paper or model.