Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Differential Impedance LNA Match AD9361

Status
Not open for further replies.

HardwareChap

Junior Member level 1
Joined
May 7, 2015
Messages
17
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,445
I'm trying to match AD9361 differential RX at 1.3 GHz. I've got the S11 parameters for the RX LNA at 1.3 GHz:

freq magS11 angS11
1300000000 0.453016055876785 -82.7105506901038

I'm matching this to the balanced input of a balun. Am I right in thinking that the way to design the balanced impedance match is to first design a standard single ended matching network and then divide the passive components by 2 to match differentially? I plugged the complex conjugate S11 parameters into a matching calculator and got :

match calc.gif

My concern is the unrealistic real world values (such as 585.3 pH). The RX pins of the AD9361 have 1.3 V on them, so need AC coupling caps to protect against DC shorts to ground. Which kind of sent me down this route:

single to diff circuit.gif

Am I right in setting the Zs impedance (balun) to 50+j0? Am I right in matching to 0.453+j82.711? Is the conversion from single ended match to diff match circuit correct?

Extra info pages 62-68 of datasheet.

- - - Updated - - -

The demo board (schematic here, sheet 2, can word search for "RX1A" to jump to relevant part of the schematic) for this RF chip doesn't have any matching implemented because of the wide range it is expected to work with, so can't really use this as a guide.
 

Your matching circuit is completely wrong because you misunderstood the impedance specification. It's 50 ohms referred S11 magnitude/phase, not real and imaginary impedance.

I suggest to use a 50 ohm Minicircuits transformer, as in the evaluation board.
 

Danke FvM. That was an obvious and silly mistake on my part! :oops: Would you have any concerns about reducing the receive sensitivity by not having any matching?
 

The datasheet is saying S11 = -10 dB without matching. That's not mind blowing but acceptable for wideband.

A simple LC network calculated for one frequency as in your example (if calculated correctly) will most likely worsen matching at other frequencies. A useful wide band matching network would involve many LC segments and a complicated design method.
 

Hi.

AD9361 Rx input impedance magnitude (differential) is a little over 100ohms at 1.3GHz (there is a table in the design guide that shows this). If you are looking to match to 50ohm characteristic impedance their best bet would be to use a 1:2 impedance transformation balun. With no DC load at Rx input (assuming you don't need a powered antenna) you would not even need AC coupling caps.

That is the simplest solution, and what we (the manufacture) would recommend. Have a look at the different baluns & AC coupling we suggest at different frequencies here:
https://wiki.analog.com/resources/eval/user-guides/ad-fmcomms2-ebz/hardware/configuration_options

Any matching attempt at 1.3GHz without including layout effects - most likely will not yield good results. In order to include layout parasitics a good simulation tool is needed. A match with a balun can be improved by adding a differential PI network between the balun and Rx input. It gives an option to empirically tune the circuit but that requires a lot of experience and skill (as well as skill with a soldering iron).

Balun insertion loss needs to be minimized to prevent sensitivity degradation. It is a dB for dB degradation. Low insertion loss baluns are readily available in that frequency range, so this is not an issue most of the time.

If you want to follow up, you can work with us directly at:
https://ez.analog.com/community/wide-band-rf-transceivers

-Robin
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top