MSG and matching - RF Cafe Forums

The original RF Cafe Forums were shut down in late 2012 due to maintenance issues - primarily having to spend time purging garbage posts from the board. At some point I might start the RF Cafe Forums again if the phpBB software gets better at filtering spam.

Below are the old forum threads, including responses to the original posts.

-- Amateur Radio
-- Anecdotes, Gripes & Humor
-- Antennas
-- CAE, CAD, & Software
-- Circuits & Components
-- Employment & Interviews

-- Miscellany
-- Swap Shop
-- Systems
-- Test & Measurement
-- Webmaster

guest
Post subject: MSG and matching Posted: Tue Jan 24, 2006 5:10 pm
hello guys

when you are designing a standard small signal amplifier you match your input/output to get as much of the MSG as possible. but what about power amplifiers?

i checked MSG of a transistor and it came out to be 13 dB at center frequency but the transistor is being used for a PA.
true or false:
-since my output is not conjugate matched but is matched for optimal load i will see much less than 13 dB of gain because of this matching (optimal laod matching).

i think the above sentence is true then my question is how can i get more gain out of the transistor (upto MSG of 13 dB) without using another stage. my guess is it is not possible considering i can't change my bias because my bias is restricted by the class of PA i am using and the amount of swing i need at the load.
thanx for help

Top

IR
Post subject: Posted: Wed Jan 25, 2006 3:31 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello guest,

Yes you are right in your description. MSG is ususally being considered for low-power amplifiers.

For PA, the matching is done per various criteria as Efficiency, Linearity or maximal output power and not necessarily for gain. The matching can be obtained by performing Load-Pull measurement. The outcome of these measurements is a set of countours on the Smith Chart.

_________________
Best regards,

- IR

Top

guest
Post subject: Posted: Mon Jan 30, 2006 6:16 am
hello guys

need to clarify about MSG and PA again. when we say that PA has gain of, let's say, 5 dB then it means Pgain = Pout - Pin = 5 dB. as i said in my earlier post the MSG of the transistor was 13 dB. my question is does MSG directly relates to Pgain---higher MSG will give me higher Pgain? in other words Pgain is directly proportional to MSG?

the thing is that my design required 15 dBm of output power with no spec for input power level so i designed it for 15 dBm of output power but i needed 10 dBm of input power to get that much output power. i am wondering if this is a poor design because i am requiring so much input power but i wasn't sure the exact relationship b/w MSG and Pgain. thanks

Top

IR
Post subject: Posted: Mon Jan 30, 2006 4:01 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello,

MSG is proportional to the Power Gain.

MSG is the maximal gain available from an amplifier without causing it to oscillate - This is the gain at which the Rollet factor K=1.

If your gain is 5dB and the MSG is 13dB, then I would say that you have some problem with your design: Mismatching in the input and/or output, wrong bias or just a failed device. The Power Gain shouldn´t be that far from the MSG.

_________________
Best regards,

- IR

Top

guest
Post subject: Posted: Mon Jan 30, 2006 11:49 pm
thanks IR; yes there is a mismatch at both the input and output because the match is for output power and not for maximum power swing--this transistor is being used for a PA. the matches were determined using source and loadpull...also the bias is not ideal for MSG either because i need certain bias for a specific class operation....you think both these factors will reduce the gain this much, from 13 dB to 5 dB? thanks

Top

rk
Post subject: Posted: Tue Jan 31, 2006 1:40 am
Matching for MSG vs. Pmax shouldn't result in 8dB difference. There's something wrong with your design.

Which doesn't neccessarily make you a bad person.

Top

IR
Post subject: Posted: Tue Jan 31, 2006 1:42 am

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Yes that could be.

I will need more information in order to help you. Please feel free to send me an email if this is more convenient . My email is: itre27@012.net.il

Thanks.

_________________
Best regards,

- IR

Top

Guest
Post subject: Posted: Wed Feb 01, 2006 2:10 am
i think your design might be right and you can get much lower than your MSG. Remember, MSG is calculated using small signal and the transistor is being used for PA and most probably you are running it at compression (not small signal) so gain will be less. On top of that, you are not cojugately matched (most probably optimal load match for max. power/efficiency, etc) so that reduces your gain too. so, i would say yeah 8 dB down from you MSG is not that odd.

-jk

Top

guest
Post subject: Posted: Wed Feb 01, 2006 7:33 am
i might have figured out the problem. why do we switch from MSG=mag(S21)/mag*S12) to MAG=mag(S21)/mag*S12)* (k-sqrt(k^2 -1)) when k>1? my k was greater than one but i was looking at MSG and not MAG. now MAG is closer to what i'm getting but i would still like to know why this switch. is just that the amplifier is now stabl (k>1) and you can get only MAG gain out of it. if that is true then i think MSG wording is confusing. i thought it meant that the circuit is stable and you can get MSG gain from it but i guess what it means is that the amplifier is conditionally stable (with the condition that you don't want to extract more than MSG).

Top

IR
Post subject: Posted: Wed Feb 01, 2006 6:00 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello,

MSG is obtained from the MAG, simply by narrowing the equation to S21/S12 when K=1 (You can try and see it for yourself).

It is a good design practice not to come close to MSG and keep some margin from this value. Now the actual gain you will get from your amplifier is dependent on the matching you provide at the input and output. As I mentioned in one of the previous posts, you can do matching according to different criteria, and for each matching criteria you will get a different gain value.

_________________
Best regards,

- IR

Top

guest
Post subject: MSG and matchingPosted: Tue Feb 07, 2006 12:23 pm
Quote:
On top of that, you are not cojugately matched (most probably optimal load match for max. power/efficiency, etc) so that reduces your gain too. so, i would say yeah 8 dB down from you MSG is not that odd.

What is the difference between being conjugately matched (max power flow) and optimal load match, if any?

Im familar with the terms of Power gain, Transducer gain and Available gain, also the MSGs/MAGs. Currently I've only been tuning so that the trade-off is between gain, noise and matching (vswrs at input, output or either one).Is Efficiency, Linearity or maximal output power mostly conciderations to take into account when dealing with large signal amplifiers?

Sincerly,
RF-student

Top

IR
Post subject: Posted: Tue Feb 07, 2006 5:10 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello,

There is no difference between the two. If you have maximal power flow that means that you are matched to the optimum.

Yes linearity, efficiency and maximum output power are related to large-signal amplifiers - Power Amplifiers.

_________________
Best regards,

- IR

Top

lifan
Post subject: Posted: Fri Feb 10, 2006 2:17 am

Captain

Joined: Fri Feb 10, 2006 1:20 am
Posts: 10
Hello,

what is MSG?

thank you.

_________________
Friends are all over the world.

Top

IR
Post subject: Posted: Sun Feb 12, 2006 4:38 am

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello lifan

MSG - Maximal Stable Gain.

It is the maximal gain that an amplifier can produce without becoming unstable (oscillate). It is given by:

MSG=S21/S12

At this Gain the Rollet factor K equals 1.

_________________
Best regards,

- IR

Top

Jeanalmira
Post subject: Posted: Tue Feb 28, 2006 5:03 am

General

Joined: Tue Mar 15, 2005 11:43 pm
Posts: 65
Location: Singapore
Hi :

I agree with IR, that the large signal gain should be close with MSG. It can be varied between 1 dB range. However, if your gain is so much off by 8 dB. I think it's probably caused by mismatch impedance of input and output.
I encountered this problem before, have you checked the return loss? I suspect maybe your output impedance too low. If there's mismatch, maybe you should check the matching for power matching.

Or you may want to check your device, whether it is in saturation region, so it affects the efficiency and linearity?

If you provice me more details, I may be able to help.

Thanks and Regards,
Jean

Posted 11/12/2012