Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

RF testing of three port device with two port VNA.

Nancy12345

Junior Member level 3
Joined
Oct 13, 2017
Messages
25
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
242
Hi,

I need help regarding testing my RF IC. I have designed a phase shifter which has differential inputs and single ended output. The VNA in my lab is a two port VNA for single ended input and output.
Is there a way to buy an input balun or if I design one on the PCB, can I use that to input differential signals in my DUT for the testing?
The operating frequency of the DUT is 8 GHz.

Please if you have experiene in RF testing, help me with tis issue.

Thank you
 

BradtheRad

Super Moderator
Staff member
Joined
Apr 1, 2011
Messages
13,891
Helped
2,757
Reputation
5,512
Reaction score
2,665
Trophy points
1,393
Location
Minneapolis, Minnesota, USA
Activity points
103,647
Your situation sounds similar to the tv antenna cable formats... one old, one new. The old 300 ohm twin lead is differential. Later came 75 ohm coax which has the shield (ground) around a center wire (signal).

A converter unit is recommended to go from one format to the other. Not sure if it's a balun or transformer. Of course tv is VHF-UHF, not 8 GHz.
 

BigBoss

Advanced Member level 5
Joined
Nov 17, 2001
Messages
4,944
Helped
1,498
Reputation
2,994
Reaction score
1,339
Trophy points
1,393
Location
Turkey
Activity points
29,816
You can simply connect two identical devices in cascade then extract s-parameters.
OR
Terminate one of the differential inputs by 50 Ohm and measure port-to-port then repeat the same for other port,After all, extract s-parameters by converting single ended to differential by aid of any RF simulator or Matlab.
 

Nancy12345

Junior Member level 3
Joined
Oct 13, 2017
Messages
25
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
242
You can simply connect two identical devices in cascade then extract s-parameters.
OR
Terminate one of the differential inputs by 50 Ohm and measure port-to-port then repeat the same for other port,After all, extract s-parameters by converting single ended to differential by aid of any RF simulator or Matlab.
Thank you so much for answering the question.
My device is a phase shifter which uses a RC poly phase filter, it uses two differential signals to generate four 90 degree apart signals. if one differential signal is not there, PPF wil not work as far as I know.
Can I just use an external balun? Are they availabe for 8 GHz?
--- Updated ---

Your situation sounds similar to the tv antenna cable formats... one old, one new. The old 300 ohm twin lead is differential. Later came 75 ohm coax which has the shield (ground) around a center wire (signal).

A converter unit is recommended to go from one format to the other. Not sure if it's a balun or transformer. Of course tv is VHF-UHF, not 8 GHz.
Thank you so much for answering the question.
Can I buy a external balun operating at 8 GHz and use it?
 

BigBoss

Advanced Member level 5
Joined
Nov 17, 2001
Messages
4,944
Helped
1,498
Reputation
2,994
Reaction score
1,339
Trophy points
1,393
Location
Turkey
Activity points
29,816
Polyphase Filter is used to obtain 4 Single-Ended Signals 90 degree apart with a single Differential Input.But you may use one output or multiple outputs.Therefore Phase Relation is very important in this case.The best way to measure the Ports one-by-one the combine the s-parameters.
Balun will not work very well because Balun will also have a Phase Shift from ideal value.
 

Nancy12345

Junior Member level 3
Joined
Oct 13, 2017
Messages
25
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
242
Polyphase Filter is used to obtain 4 Single-Ended Signals 90 degree apart with a single Differential Input.But you may use one output or multiple outputs.Therefore Phase Relation is very important in this case.The best way to measure the Ports one-by-one the combine the s-parameters.
Balun will not work very well because Balun will also have a Phase Shift from ideal value.
Ok. So terminating one load with 50 ohm and taking measurement from other and then repeat the process with changing the ports is the best way for measurement?

Thank you.
 

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
7,140
Helped
2,092
Reputation
4,188
Reaction score
1,939
Trophy points
1,393
Location
USA
Activity points
57,288
You can get baluns but putting them in the
lineup adds the balun frequency response to
the mix.

I imagine you'd first characterize the balun
with the VNA, loaded as you'd use it, and then
pull the device data from which to de-embed
the balun's contribution.
 

Nancy12345

Junior Member level 3
Joined
Oct 13, 2017
Messages
25
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
242
You can get baluns but putting them in the
lineup adds the balun frequency response to
the mix.

I imagine you'd first characterize the balun
with the VNA, loaded as you'd use it, and then
pull the device data from which to de-embed
the balun's contribution.
Thank you so much for answering the question.

Do I need to buy some 50 ohm termination loads as well for proper calibration?
 

ktr

Full Member level 3
Joined
Jun 11, 2018
Messages
152
Helped
15
Reputation
30
Reaction score
15
Trophy points
18
Location
Turkey
Activity points
1,230
You can do both and compare the results honestly. Solution Bigboss offers is a solid one. I would personally do 2 different measurements and compare them to each other.

1- Solution of Bigboss

2- get a https://www.minicircuits.com/WebStore/dashboard.html?model=TCM1-83X+ ; I used this balun up to 8 Ghz. It works nicely. Get some 50 ohm terminations; which should have Return loss better than at least -30db.

Compare the two results; with the difference between them equaling balun's added noise. You can also remove the balun's effect if you can add it to your calibration data; but that is a easier said than done in practice.
 

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top