Measuring multiport devices with a two-port VNA with ...

Measuring multiport devices with a two-port

VNA with traceability to national standards

Martin Salter

26th ANAMET meeting 18th September 2006

Overview

• Measurement Method for 3-ports

• Measurement of a 3-port directional coupler

• Measurement of a 3-port power splitter – with a short-circuit as termination

3-port directional coupler

Port 1

(female) Port 2

(male)

Port 3

(female)

Terminating

load

(permanently

attached)

Coupler measurands:

Coupling factor (dB)

Directivity (dB)

Magnitude of S11, S22 & S33

Measurement method (1)

• Method due to Speciale (MTT-30, 1982)

• Calibrate VNA with two male test ports (use LRL with female-female lines)

• Measure female-female adapter

Measurement method (2)

• Measure male and female loads

• De-embed adapter

VNA

Port 1

(male)

VNA

Port 2

(male)

(*) Adapter

(female-female)

Terminating

load

(male)

Terminating

load

(female)

Measurement method (3)

• Measure three partial 2-ports

(with the unused

port terminated with one of the

measured loads)

• De-embed adapters

VNA

Port 2

(male)

VNA

Port 1

(male)

(*) Adapter

(female-female) 1 2

3

Terminating

load

(male)

VNA

Port 1

(male)

VNA

Port 2

(male)

Ad

apte

r (*

)

(fem

ale

-fem

ale)

1

3

2

Terminating

load

(male)

VNA

Port 1

(male)

VNA

Port 2

(male)

1

2

3

Terminating

load

(female)

Measurement method (4)

• Renormalise scattering matrix of partial 2-ports to ZM at a male port and ZF at a female port

• Assemble 3-port scattering matrix normalised to ZM at male ports and ZF at female ports

• Renormalise 3-port scattering matrix to 50 ohms at all ports

• Calculate measurands from 3-port S-parameters

• Use a Monte Carlo method to propagate uncertainty from raw S-parameter measurements through adapter de-embedding and impedance renormalisation to measurands

Measurement of a 3-port directional coupler (1)

0.0000

0.0020

0.0040

0.0060

0.0080

0.0100

0.0120

0.0140

0.0160

0.0180

150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

Frequency (MHz)

Lin

ea

r m

ag

nit

ud

e S

11

Measurement of a 3-port directional coupler (2)

0.0000

0.0020

0.0040

0.0060

0.0080

0.0100

0.0120

0.0140

0.0160

0.0180

150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

Frequency (MHz)

Lin

ea

r m

ag

nit

ud

e S

22

Measurement of a 3-port directional coupler (3)

0.0000

0.0020

0.0040

0.0060

0.0080

0.0100

0.0120

0.0140

0.0160

0.0180

150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

Frequency (MHz)

Lin

ea

r m

ag

nit

ud

e S

33

Measurement of a 3-port directional coupler (4)

19.000

19.500

20.000

20.500

21.000

21.500

22.000

22.500

23.000

23.500

24.000

150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

Frequency (MHz)

Co

up

lin

g f

ac

tor

(dB

)

Measurement of a 3-port directional coupler (5)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

Frequency (MHz)

Dir

ec

tiv

ity

(d

B)

Measurement of a 3-port power splitter – with a short-circuit as termination (1)

• Renormalisation corrects for mismatch of terminating load

• If load is nominally matched correction should be small

• But what if termination is a high reflect e.g. a short-

circuit?

• Test of method

Measurement of a 3-port power splitter – with a short-circuit as termination (2)

1 2

3

VNA

PORT 1

VNA

PORT 1 VNA

PORT 1 VNA

PORT 2

VNA

PORT 2

VNA

PORT 2

1

2 3 1

2

3

First partial

two-port

Second partial

two-port

Third partial

two-port

Terminating load is either

nominally matched or else a

short-circuit

Power splitter measurands:

Input VSWR, effective output VSWR,

Output tracking (dB)

Measurement of a 3-port power splitter – with a short-circuit as termination (3)

0.94

0.95

0.96

0.97

0.98

0.99

1.00

1.01

1.02

1.03

1.04

1.05

0 5 10 15 20 25 30

Frequency (GHz)

Mag

nit

ud

e S

ho

rt-C

ircu

it V

RC

Measurement of a 3-port power splitter – with a short-circuit as termination (4)

-180

-135

-90

-45

0

45

90

135

180

0 5 10 15 20 25 30

Frequency (GHz)

Ph

as

e s

ho

rt-c

irc

uit

VR

C (

de

gre

es

)

Measurement of a 3-port power splitter – with a short-circuit as termination (5)

1

1.1

1.2

1.3

1.4

1.5

1.6

0 5 10 15 20 25 30

Frequency (GHz)

Inp

ut

VS

WR

at

po

rt 1

Short circuit

Matched load

Measurement of a 3-port power splitter – with a short-circuit as termination (6)

1

1.1

1.2

1.3

1.4

1.5

1.6

0 5 10 15 20 25 30

Frequency (GHz)

Eff

ec

tiv

e o

utp

ut

VS

WR

at

po

rt 2

SC

Matched load

Measurement of a 3-port power splitter – with a short-circuit as termination (7)

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

0 5 10 15 20 25 30

Frequency (GHz)

Eff

ec

tiv

e o

utp

ut

VS

WR

at

po

rt 3

Short circuit

Matched load

Measurement of a 3-port power splitter – with a short-circuit as termination (8)

-1.5

-1

-0.5

0

0.5

1

1.5

0 5 10 15 20 25 30

Frequency (GHz)

Ou

tpu

t T

rac

kin

g (

dB

)

Short circuit

Matched load

Conclusion

• Results presented for a directional coupler and a power splitter measured with a 2-port VNA

• Correction applied for mismatch of terminating load

(renormalisation)

• Correction still works if terminating load is a short-circuit

• Forms the basis of an NPL 3-port calibration service