•−•  ••−•    −•−•  •−  ••−•  •
RF Cafe Morse Code >Hear It<

Job Board

About RF Cafe™

Sitemap

Matching network for balanced RF inputs? - RF Cafe Forums

Because of the high maintenance needed to monitor and filter spammers from the RF Cafe Forums, I decided that it would be best to just archive the pages to make all the good information posted in the past available for review. It is unfortunate that the scumbags of the world ruin an otherwise useful venue for people wanting to exchanged useful ideas and views. It seems that the more formal social media like Facebook pretty much dominate this kind of venue anymore anyway, so if you would like to post something on RF Cafe's Facebook page, please do.

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


MRW
Post subject: Matching network for balanced RF inputs? Posted: Wed Jan 18, 2006 10:18 am

Captain

Joined: Wed Jan 18, 2006 9:46 am
Posts: 5
Location: Earth
Hi!

I am messing around with the TEA5767 (FM receiver IC). I am trying to figure out how they came up with the following matching network:



It's equivalent circuit looks like this:



In the application, note it states that "the antenna-input signal is fed into the balanced FM-RF inputs (pin 28 and pin 30) via a RF matching impedance circuit. A series capacitor, two parallel capacitors and a coil (L1) build this circuit.

Also, it states that the "TEA5767HN has an integrated low noise amplifier (LNA). This is a balanced amplifier, which is less sensitive for common mode noise, compared with a single ended LNA. The input impedance of the LNA is (100 ohm || 4pF) each pin (referenced to ground).

Any tips?

I tried designing the matching network for 88 MHz, so I got 95.337 - j21.085 for my input impedance, but I can't seem to get the values that they have using a 40ohm source.

Thanks!
[/img]

_________________
Don't hate me because I'm a newbie.


Top

IR
Post subject: Posted: Wed Jan 18, 2006 5:12 pm

Site Admin


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

The motivation of using a balanded LNA is due to the fact that FM transmission is more susceptible to noise because of its larger BW.

To find out the values of the matching components for 88MHz, you should first know the impedance of your antenna at this frequency.

It will help if would know the frequency used for these values. If you will know the impedance of your antenna and thus the location of it on the Smith Chart and you already know the input impedance to the LNA inside the IC, you can easily simulate a matching circuit - you even have the structure of the network already.

_________________
Best regards,

- IR


Top

Guest
Post subject: Balanced inputPosted: Wed Jan 18, 2006 8:38 pm
The usual FM broadcast antenna for portable receivers (which I believe is the target market for the TEA5767) is a shortened dipole - less than a half-wave long. This antenna will look (mostly) capacitive. If you assume that it looks like j40 Ohms, you get close to 36 pF around 108 MHz. That 36 pF in series with the 100 pF comes out to about 26 pF. 26 + 27 pF = 54 pF, reasonably close to the 47 pF on the other side of ground in the tank.

So I don't think Philips had anything particularly technically high-powered in mind here - just a tank resonant in the FM band, with a center-tapped capacitor to do the conversion to balanced form.

Good Luck!


Top

Guest
Post subject: Posted: Thu Jan 19, 2006 10:07 am
IR wrote:
To find out the values of the matching components for 88MHz, you should first know the impedance of your antenna at this frequency.

It will help if would know the frequency used for these values. If you will know the impedance of your antenna and thus the location of it on the Smith Chart and you already know the input impedance to the LNA inside the IC, you can easily simulate a matching circuit - you even have the structure of the network already.


Thanks, IR! I was just assuming that the antenna impedance would be just around 40 ohms since they said that it is a 40 ohm antenna. So I just tried to match the LNA input impedance to 40 ohms.

Also, I think my main confusion lies in the structure of the matching network because it states that the two capacitors are in parallel, but that would make the inductor a short (?).


Top

MRW
Post subject: Re: Balanced inputPosted: Thu Jan 19, 2006 10:14 am

Captain

Joined: Wed Jan 18, 2006 9:46 am
Posts: 5
Location: Earth
Guest wrote:
The usual FM broadcast antenna for portable receivers (which I believe is the target market for the TEA5767) is a shortened dipole - less than a half-wave long. This antenna will look (mostly) capacitive. If you assume that it looks like j40 Ohms, you get close to 36 pF around 108 MHz. That 36 pF in series with the 100 pF comes out to about 26 pF. 26 + 27 pF = 54 pF, reasonably close to the 47 pF on the other side of ground in the tank.

So I don't think Philips had anything particularly technically high-powered in mind here - just a tank resonant in the FM band, with a center-tapped capacitor to do the conversion to balanced form.

Good Luck!


Thanks, Guest! I am not quite that familiar with antennas, so I just assumed the antenna would be just 40 ohm. I guess I'm still seeing matching networks matched to a 50ohm source or load.

Does the 40ohms antenna impedance refer to the impedance I need to use to normalize my other impedance values for use on a Smith Chart?

Thanks!

_________________
Don't hate me because I'm a newbie.


Top

MRW
Post subject: Re: Balanced inputPosted: Thu Jan 19, 2006 10:28 am

Captain

Joined: Wed Jan 18, 2006 9:46 am
Posts: 5
Location: Earth
Guest wrote:
.... That 36 pF in series with the 100 pF comes out to about 26 pF. 26 + 27 pF = 54 pF, reasonably close to the 47 pF on the other side of ground in the tank.


By the way, how is 26pF in parallel with 27pF?

Thanks! I would like to understand this matching configuration more.

_________________
Don't hate me because I'm a newbie.


Top

IR
Post subject: Posted: Thu Jan 19, 2006 3:35 pm

Site Admin


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

It can be said that the capacitors are connected in parallel from a DC point of view, because in DC the inductor is considered as short.
From AC point of view there is a tank circuit for each of the balanced inputs consisting from each capacitor and the inductor.

Hope this helps!

_________________
Best regards,

- IR


Top

MRW
Post subject: Posted: Thu Jan 19, 2006 5:18 pm

Captain

Joined: Wed Jan 18, 2006 9:46 am
Posts: 5
Location: Earth
Thanks again, IR!

I'm still trying to understand it more. I guess since I'm no expert, then I would need to see the matching network design for the specific IC from scratch.

I also tried redrawing the matching network to see if I could get a different perspective. But now I'm just stuck looking at it and the ZY chart in an empty manner.

Redrawing matching network:


Frequency response of matching network from application notes:


My gratitudes to additional insight.

_________________
Don't hate me because I'm a newbie.


Top

IR
Post subject: Posted: Thu Jan 19, 2006 6:09 pm

Site Admin


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

Your drawing is correct.

The output response also makes sense: At the frequency of operation you have minimal insertion loss, that means there is a good impedance matching between the IC's LNA to the antenna. Out of the band the insertion loss decreases, which means that the matching is degraded and that is exactly what you want to achieve: A selective narrowband matching network which provides good matching to the antenna only around the BW of operation.

You can get more insight if you will view the Smith Chart or the return loss graph.

Don't hesitate to ask, that is the best way to learn!
No one is born RF Engineer

_________________
Best regards,

- IR


Top

Guest
Post subject: Posted: Thu Jan 19, 2006 7:14 pm
The circuit is based on a transmission line balun, where a half wave tx line is placed between RFin1 and RFin2. This inverts the drive phase at RFin2 and parallels the individual single ended impedances when looking into RFin1 and the line. This is what gives the apparent transform by going from two impedances in series to two in parallel. Your circuit minus the 100pf blocking cap is a lumped approximation of a half wave tx line when all of the shunt caps are accounted for.

The basic circuit is only meant to do a 4 to 1 transform. You can increase the transform by starving the DC blocking cap and then playing with the other values. It is important to check the common mode effects after doing this.


Top

Honza
Post subject: Posted: Fri Jan 20, 2006 2:28 am

Captain

Joined: Tue Dec 13, 2005 2:42 am
Posts: 5
Location: Czech Rep.
...One thing is still confusing me. Why the manufacturer in his datasheet uses balanced RF input for UNBALANCED antenna?

Someone of you mentioned that balanced input is used in order to minimize common-mode noise. It make sense. But if you would like to minimize the common-mode noise, you must have ALL the receive chain balanced?! Isn't it? ...It's not sufficient to have just only a part of the chain balanced! Am I right? ...Because unbalanced antenna (as it was said) is less sensitive to the common-mode noise so it receives "a lot of" "single-mode" noise and converts this noise into the common-mode noise. So balanced RF input for unbalanced antenna is useless, am I right?

In other words by me it's useless to mix balanced and unbalanced elements in one chain (looking on the immunity against noise and not on matching).

What do you think about it?

Honza.


Top

MRW
Post subject: Posted: Wed Jan 25, 2006 12:12 pm

Captain

Joined: Wed Jan 18, 2006 9:46 am
Posts: 5
Location: Earth
Thanks, all! By the way, I'm curious about the balance inputs of the LNA. What provides the 180 phase shift for the RF signal? Does the matching network do this?

_________________
Don't hate me because I'm a newbie.


Top

malvinas2
Post subject: Posted: Fri Jan 27, 2006 5:14 am
MRW, I don't know how to describe the matching network in analytical terms, but to see the phase shift of 180° it would be sufficient to simulate it with ADS or something like that. The difference in phase between S_RFI1 and S_RFI2 will result in 180°.

BTW: Why do you choose 88 MHz ? If FM band ranges from 88 to 106 MHz wouldn't it be better do choose center frequency, 102 MHz ?



Posted  11/12/2012
Custom Search
More than 10,000 searchable pages indexed.

Your RF Cafe
Progenitor & Webmaster

Click here to read about RF CafeKirt Blattenberger... single-handedly redefining what an engineering website should be.

View the YouTube RF Cafe Intro Video Carpe Diem!
(Seize the Day!)

5th MOB: My USAF radar shop

Airplanes and Rockets: My personal hobby website

Equine Kingdom: My daughter Sally's horse riding website