The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Mar 29th, 2024, 2:20am
Pages: 1
Send Topic Print
Power vs conjugate matching at SAW filter input in TX chain (Read 8816 times)
Mirko
New Member
*
Offline



Posts: 5

Power vs conjugate matching at SAW filter input in TX chain
Sep 01st, 2014, 7:13am
 
Hello to everyone,

this is my first post after some time of lurking, so I'd like to start by thanking the contributors to this forum for the useful information I've already found here. Even though similar topics have been discussed in the past, I was unable to find an answer to the following; hope this is not a repost, or too a trivial question.
I am currently designing an OET module whose transmitter path is composed by:
1) the integrated PA of the transceiver
2) DC block and a discrete, L matching network
3) a SAW filter, whose ports are nominally at 50 Ohm, to attenuate spurious before boosting the signal, and
4) an external front-end module (basically a constant-gain booster in its TX section), 50 Ohm input too, followed by the antenna matching.
Initially, I was supposed to realize a power match (as opposed to a conjugate one, borrowing the terminology from dr. Cripps' book) at the transceiver output; but does this actually make sense? My doubts come out of two main reasons: first, I fear the mismatch at the input of the SAW filter would negatively affect its passband flatness; and additionally, a standing wave would be present between the two stages.
My understanding was therefore that in such a situation (cascaded - discrete - stages with a SAW filter in between), a conjugate match should be preferred. Do you think this reasoning is flawed in some way?

Thank you
Regards
Back to top
 
 
View Profile   IP Logged
aaron_do
Senior Fellow
******
Offline



Posts: 1398

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #1 - Sep 4th, 2014, 5:37am
 
Hi,


I think nobody responded because you already seem to know the answers, but you just want somebody to confirm your decision. Since it is just an interface between two of your own components (as opposed to providing a source to the outside), I think you only need to work in terms of the effect on the overall system. i.e. make sure the overall flatness is good, the harmonics are good etc.

In terms of the standing wave, it will depend on frequency and the length of your transmission line. For the gain flatness and roll-off of the filter, your best bet is to run a simulation.

btw, what is an OET module?


regards,
Aaron
Back to top
 
 

there is no energy in matter other than that received from the environment - Nikola Tesla
View Profile   IP Logged
Mirko
New Member
*
Offline



Posts: 5

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #2 - Sep 4th, 2014, 7:15am
 
Hello Aaron,
thank you for your reply.
I was actually quite unsure, mainly because I fear my theory may miss some important point. Being a board-level designer, I guess I am probably spoiled by the many internally-matched PAs on the market!

Quote:
btw, what is an OET module?  

Sorry, my (double) mistake. It should have been OEM, Original Equipment Manufacturer, in the sense that the module will use third part ICs (I guess it is also called Value-Added Reseller, but maybe I'm just misusing the expression); admittedly, quite a useless information in this thread. Seems like I'd better re-read before posting!

Regards
Mirko
Back to top
 
 
View Profile   IP Logged
wave
Senior Member
****
Offline



Posts: 117
Silicon Valley
Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #3 - Sep 4th, 2014, 10:48am
 
Mirko wrote on Sep 1st, 2014, 7:13am:
I was supposed to realize a power match (as opposed to a conjugate one, borrowing the terminology from dr. Cripps' book) a


Since I don't have the text, can you elaborate when is a power match not achieved by a conjugate match??    :-?
Back to top
 
 
View Profile   IP Logged
aaron_do
Senior Fellow
******
Offline



Posts: 1398

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #4 - Sep 4th, 2014, 5:01pm
 
Hi,


my understanding. Power Match, Zsource = Zload*. Impedance Match, Zsource = Zload.

I'm getting a little confused though now. So my understanding is that the SAW filter is not quite 50-ohm (there is some imaginary part), and you need to decide whether to do an impedance match or a power match. The SAW filter is probably designed for 50-ohm terminations anyway, but I'm thinking that if you're worried about reflection, then the important impedance is the characteristic impedance of the transmission line, right? If your transmission line is short, then reflection doesn't come into the equation anyway...


Aaron
Back to top
 
 

there is no energy in matter other than that received from the environment - Nikola Tesla
View Profile   IP Logged
Mirko
New Member
*
Offline



Posts: 5

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #5 - Sep 5th, 2014, 12:08am
 
Hi,

Quote:
Since I don't have the text, can you elaborate when is a power match not achieved by a conjugate match??    :-?

As I understood it (please correct me if I'm wrong), conjugate match (or gain match) leads to maximum power transfer, but this does not imply maximum output power in a real device, since it is possible that the maximum voltage rating on the drain of the output transistors will be reach for Iout lower than the maximum current the device can source to the load.
A power match, as Cripps calls it, would instead be based on load-pull contours, so as to present the amplifier with an "optimum" load impedance which maximazes Pout for the given Iout, while at the same time keeping Vdrain to a safe value. A brief discussion on this subject, but using a slightly different terminology, can be found also on this appnote http://www.silabs.com/Support%20Documents/TechnicalDocs/AN369.pdffrom SiLabs.
Finally, setting Zload = Zsource would yield a reflectionless match.
Back to top
 
 
View Profile   IP Logged
Mirko
New Member
*
Offline



Posts: 5

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #6 - Sep 5th, 2014, 12:40am
 
Hi Aaron,
you are correct, the SAW filter is meant to connect to 50Ω, but its input impedance has an imaginary part, and moreover changes quite a lot in its passband (about two complete turns around the center of the Smith chart). The consequence of this is that, although each stage has a very good flatness taken alone, the entire transmitter chain shows variable power across the band (a similar situation is described here:http://www.maximintegrated.com/en/app-notes/index.mvp/id/3040 , the only difference being that my transceiver is a high-impedance one instead of an internally-matched one).
I actually never though about attempting a reflectionless match, I just wanted to reduce the VSWR to an acceptable level. So my doubt was, since the SAW filter (despite its twisting S11) is design to have a 50Ω input, am I correct in assuming that giving it a different impedance will in general worsen its flatness? If so, power matching is not an option, since it would imply (looking towards the generator) transforming the SAW filter input impedance to the optimum one.
Your observation regarding transmission line length makes perfect sense; I didn't mention that on the SAW input side, the length is negligible. On its output side, that is, between SAW filter and PA, it is less than a tenth of the wave length.
Hope I've not been too confused
Thank you

Regards
Mirko
Back to top
 
 
View Profile   IP Logged
aaron_do
Senior Fellow
******
Offline



Posts: 1398

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #7 - Sep 5th, 2014, 9:04am
 
Hi Mirko and Wave,


sorry but I misunderstood the original question. I didn't remember the terminology used in Cripp's book and assumed that by power match vs. impedance match you were referring to maximum power transfer vs. minimum reflection. Now I realise you are talking about maximum output power vs. minimum reflection.

So what you're saying is the SAW is designed to have a 50-ohm source impedance, but the PA's optimum output impedance for power/efficiency is something else (probably high impedance).

In this case, I think its gonna have to come down to simulation. If you go with a power match, you need to make sure you can still keep your harmonics low at the final output. This might be the more manageable solution. If you go with an impedance match, you could potentially run into a lot of other problems. 1) You will have to drive the PA Driver harder which will degrade the linearity. 2) Efficiency of the PA Driver will probably degrade too.

My thinking is that the PA Driver is designed to drive a 50-ohm load (or whatever the datasheet says) for the optimal trade-off between linearity and efficiency. Furthermore, Neither the gain flatness nor the actual harmonics coming out of the PA Driver is necessarily better with an impedance match as it strongly depends on the internal designs of the individual components. I would simulate it, but I'm guessing power match (50-ohm) is the correct answer. Just my thinking.


regards,
Aaron
Back to top
 
 

there is no energy in matter other than that received from the environment - Nikola Tesla
View Profile   IP Logged
Mirko
New Member
*
Offline



Posts: 5

Re: Power vs conjugate matching at SAW filter input in TX chain
Reply #8 - Sep 6th, 2014, 11:41am
 
Hi Aaron,

thank you for taking the time to providing valuable suggestions; the confusion is probably my fault, since it seems that the terminology is far from being universally accepted (in the literature "power match" is used with both meanings depending on the source, as far as I can tell), that's why I mentioned the author, but it would have been clearer to just explain what I meant in plain words.
Just to be sure:
Quote:
Now I realise you are talking about maximum output power vs. minimum reflection.
,
--> maximum output power vs. 50Ω match.

Unfortunately, I don't have detailed description of the internal circuitry of the PA driver, therefore I cannot simulate it; all things considered, I think I will go for a 50Ω match at the output of the PA driver (that seems to be your suggestion too, if I got it right), possibly iterativelly adjusted to correct the overall flatness.

Quote:
Furthermore, Neither the gain flatness nor the actual harmonics coming out of the PA Driver is necessarily better with an impedance match as it strongly depends on the internal designs of the individual components.

sorry, once again you are perfectly correct, but my sentence was not clear: I meant that power (loadline) match would have affected the SAW filter's flatness. I just verified this yesterday, by observing that the S11 of the SAWF is strongly degraded when a complex impedance is presented at its output, instead of the prescribed 50Ω.

Thanks again for your help
Regard
Mirko
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.