Coaxial connector effects

Federal Department of Justice and Police FDJP

Federal Office of Metrology METAS

Coaxial connector effects

EURAMET RF&MW expert meeting (Torino) Juerg Ruefenacht

Johannes Hoffmann 24 May 2011

© METAS Juerg Ruefenacht + Johannes Hoffmann EURAMET TC-EM SC-RF&MW Experts meeting 23. – 24. May 2011 slide 2


2.4 mm Load (female) SN0407

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S11

(lin

)

EUROMET.EM.RF-S16 S-Parameter in 2.4 mm

12 Participants




0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00

Frequency (GHz)

S11

(lin

) DUT

Ref. plane

TP

TP SL Cal.

Meas.

pin-depth:

0.0000 mm

First investigations in 2004




0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00

Frequency (GHz)

S11

(lin

)

First investigations in 2004

DUT

Ref. plane

TP

TP SL Cal.

Meas.

SL pin-depth:

0.0000 mm

Protrusion

SL pin-depth:

+0.0051 mm



Conclusions for the found S11 variations (in 2005)

• Impact to the participants:

• Reproducibility versus traceability to SI

• One NMI even withdraw their results > 8 GHz

• S11 differences larger than stated uncertainties

• LRL : issue with pin gap (beadless centre conductor)

• Effects from connectors are dominant

• SOLT : good reproducibility (traceable to Agilent)

• Connector orientation data (repeatability)



Topics

• Round table discussion

• Measurement examples

• Change of paradigm: the out-dated dogmas

• The concept of the half connector

R&D projects: - CoMo70

- CalCon



The concept of the half connector

The half connector concept allows a consistent definition of the

reference plane and an increase in accuracy if compared to ignoring

connector effects.



MOTIVATION: consistent S-Parameter definition

SUHNER RF connector guide (2. Edition 1997)

Characterizing

microwave devices with

S-parameters requires

the definition of

S-parameters in

reference plane(s).

Coaxial components:

Reference plane in

the connector -

in the middle of a

discontinuity!

Electrical

Reference Plane

R&D projects: - CoMo70

- CalCon



S-Parameters describe electromagnetic fields

Maxwellian fields

Animation: Pascal Leuchtmann ETHZ

Transmission line model and thus S-parameters

require the definition of reference planes in the

electromagnetic field.

A reference plane is a plane which is orthogonal

to the direction of transmission.

In metrology we typically use the pseudo

S-parameter definition (monomode):

- Plain waveguide

- Fundamental mode (TEM mode)



Issue: the reference plane is located in a discontinuity Discontinuities provoke higher modes -> would require multimode S-Parameters

2.4 mm(f) Cal Standard 2.4 mm(m) Test Port

Electrical reference plane



Old dogma

still widely used at the NMI and manufacturer level:

Connectors have to be treated as a pair only !



Solution: The concept of the half-connectors

Electrical

Reference Plane

METAS MetInfo 3/2010: Mating Habits

1. step: complete connector



The female half-connector

Electrical

Reference Plane METAS MetInfo 3/2010: Mating Habits

2. step: perfect male connector



The male half-connector

Electrical

Reference Plane METAS MetInfo 3/2010: Mating Habits

3. step: perfect female connector



Connector effects

• 2.4 mm CMC entries are in the order of | S11 | = 0.020 (@ 50 GHz)

2.4 mm connector effects are in the same order!

• For a typical 1.85 mm connector the S-parameter of a

complete pair is in the order of | S11 | = 0.014 (at 67 GHz)

• For a typical 1.85 mm connector the S-parameters of a complete

pair and the cascaded S-parameters differ by about:

| S11 – S11num | = 0.00033 (at 67 GHz)

Perfect

Load S11 S11num

Perfect

Load + + +



Example 1: Air Line without connector effects

a00

a01

a10

a11

Air Line



Example 2: Air Line model with connector effects

c00

c01

c10

c11 d00 d11

d10

d01

b00 b11

b01

b10

Air Line Con. Connector



Concepts to ensure S-parameter consistency

• Physical behaviour of passive standards • All passive devices : fulfil passivity conditions

• e.g.: short and flush short : reflection coefficients not larger than 1

• e.g.: flush short : no positive phase behaviour (≤180 deg)

• e.g.: air line : no ripple on transmission coefficients

• Consistent results by using different cal methods • e.g.: S11 and S22 measurements of low loss components

• Measurement comparisons ? ? ?



● ● Short

Open

Concepts to ensure S-parameter consistency

• Physical behaviour of passive standards • All passive devices : fulfil passivity conditions

• e.g.: short and flush short : reflection coefficients not larger than 1

• e.g.: flush short : no positive phase behaviour (≤180 deg)

• e.g.: air line : no ripple on transmission coefficients



Measurement examples

Issue: cal kit standard definitions do not account for the male and

female connector effects!



Measurement of a 2.4 mm Air Line

• Air Line: length = 17 mm, beadless, slotless

• Test Port 2: slotless



Reflections of the 17 mm Air Line (SOLT calibration using the 85056A cal kit definitions)

• S11 dominant reflection at Port 2 !




Reflections of the 17 mm Air Line (SOLT calibration using the 85056A cal kit definitions)





Reflections of the 17 mm Air Line (SOLT calibration with connector effects)

• S11 dominant reflection at Port 1

• S22 dominant reflection at Port 1



2.4 mm male flush short standard



2.4 mm female flush short standard



2.4 mm Flush Short (female) SN0152

0.990

0.995

1.000

1.005

1.010

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S1

1 m

ag

(lin

)

OSL (85056A cal kit definitions)

Flush short: S11 magnitude (OSL with cal kit definitions)




0.990

0.995

1.000

1.005

1.010

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S1

1 m

ag

(lin

)


OSL (including connector effects)

Flush short: S11 magnitude (OSL including connector effects)




177.000

178.000

179.000

180.000

181.000

182.000

183.000

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S1

1 p

ha

se

(d

eg

)

OSL (generic definitions)

OSL (SL with connector effect)

LRL Multiline (incl. connectors)

Flush short: S11 phase (OSL with cal kit definitions)




177.000

178.000

179.000

180.000

181.000

182.000

183.000

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S1

1 p

ha

se

(d

eg

)


OSL (including connector effects)

Flush short: S11 magnitude (OSL including connector effects)




0.000

0.010

0.020

0.030

0.040

0.050

0.060

0.070

0.080

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

(S1

1 O

SL

85056A -

S1

1 O

SL

wit

h c

on

necto

r)

ma

g

OSL (85056A) - OSL (with connector effects)

Flush short: complex difference of the S11 measurements



2.4 mm Airline 50 ohm SN0694 (one bead)

-0.400

-0.350

-0.300

-0.250

-0.200

-0.150

-0.100

-0.050

0.000

0 5 10 15 20 25 30 35 40 45 50

Frequency (GHz)

S2

1 (d

B)

SOLT (85056A cal kit definitions)

SOLT (including connector effects)

Airline from 85057B verification kit: S21 magnitude ripple



Change of paradigm: the out-dated dogmas

The revealed systematic connector effects can no longer be ignored



Out-dated connector dogmas

• Interconnects have always to be treated as a pair

• Mind the gap

• Cal kit definitions include the connector effects

• Slotless connectors are always better





• Mind the gap





85054B Type-N Cal kit standard definition table



85056A Cal kit standard definition table (male = female)



Sliding load: connector assumed to be perfect

● Load

Im

Re

Actual

directivity S11A of load housing (beadless air line part)

S11A of the load element

Residual directivity





• Mind the gap





Example: The standard definitions do not take into account

for any connector effects


- connector

effects ignored

Connectors have always to be treated as a pair



Dogma: the same design is used for cal standards and DUTs

therefore the connector effects are calibrated out

2.4 mm(m) Test Port

DUT has the same effect!

- connector

effects ignored


2.4 mm(f) DUT



2.4 mm(m) Test Port

DUT has the same effect!

- connector

effects ignored

2.4 mm(f) Cal Standard




The electrical reference plane is not correctly defined:

effects of female connector absorbed in the test port error box



- connector

effects ignored

Issue: if treated as a pair only



Only the error introduced by the TP will be „absorbed“

into the error box of the calibration model of the VNA


- Slots accounted

- Pin-depth acc.

Electrical reference plane Defined: gives consistent results





• Mind the gap






2.4 mm(f) Test Port 2.4 mm(m) Component

pin-gap

- Typical connection

- Pin-depth on both sides





Old dogma: - Mind the pin-gap

- Ideal connection (50 ohm)

„Nominal Test Port“





Old dogma: - TP with no pin-depth

- always minimise the gap

„Nominal Test Port“



1.85 mm connector: S11 with small female chamfer

CoMo70 outcomes: 0 10 20 30 40 50 60 70 80

0

0.01

0.02

0.03

0.04

0.05

0.06

Frequency [GHz]

|S1

1|

PinGap=0.001mm

PinGap=0.005mm

PinGap=0.01mm

PinGap=0.02mm

PinGap=0.05mm

PinGap=0.1mm

Johannes Hoffmann, ETHZ, CoMo70 project (Paper published)



1.85 mm connector: S11 with big female chamfer

Issue: near field coupling effects

0 10 20 30 40 50 60 70 800

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Frequency [GHz]

|S1

1|

PinGap=0.001mm

PinGap=0.005mm

PinGap=0.01mm

PinGap=0.02mm

PinGap=0.05mm

PinGap=0.1mm

CoMo70 outcomes:

Johannes Hoffmann, ETHZ, CoMo70 project (Paper published)



Typical coupling effects when measuring beadless air lines



Old paradigm: keep pin-gaps as small as possible




CONTROL THE GAP



• Goals for the gap:

neither too small nor too large

but not as small as possible!

• The pin-gap needs to have a certain size to ensure

reproducible S-parameters. This will avoid the

unpredictable coupling effects.

• The pin-gap must be realized on the test port side.

• The smallest admissible pin-gap is different for

each connector design.

New pin-gap dogmas in RF & MW metrology:





• Mind the gap





- Slotted female CC

- Problem: pin diameter

The impedance characteristics of the connected calibration

standard is changed by the size of the TP pin diameter


„Pin diameter variations“ - „model parameter

no longer correct“



- Slotted female CC


The impedance characteristics of the connected calibration

standard is changed by the size of the TP pin diameter


- „model parameter

no longer correct“ „Pin diameter variations“



- Slotless female CC


The slotless design is more robust against pin diameter

variations: female cal standard characteristic will not change


„No nominal pin diameter“ „slotless design“



- Slotless female CC


The slotless design is more robust against pin diameter

variations: female cal standard characteristic will not change


„No nominal pin diameter“ „slotless design“



Slotted versus slotless female connector design

HP Microwave Connector Care (08510-90064)

Spring forces inner

contact forward Inserted male plug

Detail of precision slotless female centre conductor developed by Agilent

(Slotless female conductors are available for Type-N, 3.5 mm and 2.4 mm connectors)



Slotted versus slotless connector interface (Type-N)

Slotted (2-, 4-, 6- or 8-slots) Slotless (precision)

METAS Blair Hall, MSL



Slotless female connector design examples: 3.5 mm

3.5 mm contact developed by Agilent WSMA contact developed by Anritsu



2.4 mm (female) connector interface designs:

Slotted (4-slots) Slotless (precision)

Blair Hall, MSL Blair Hall, MSL



2.4 mm (f) slotless connector section from a sliding load



2.4 mm (f) slotless connector section from a sliding load



2.4 mm (female) slotless connector interface design:




„pin diameter variations“




„pin diameter variations“



• There is no better or worse – they are just different !

• Slotted connectors are more sensitive to male pin

diameter variations compared to an optimised slotless

connector design.

• Slotless connectors may have a higher reflection

coefficient compared to an optimised slotted connector

design.

• Slotless connectors can be characterised more

accurately and are, therefore, the first choice for

metrology applications.

Are Slotless connectors always better ?



Conclusions : “consistent S-parameters”

• Change of paradigm:

• Half connectors instead of pair only or ignoring them.

• Include the connector effects in the standard definitions.

• Control the gap (a finite pin-gap is a must).

• Dissemination:

• NMI – accredited laboratories – industry.

• Cal kit manufacturers.

• Traceability to SI - consistent S-parameter results.

• Taking systematic connector effects into account is the

key factor to increase the S-Parameter accuracy.



Thank you very much for your attention !

More information: www.metas.ch/hf

Federal Department of Justice and Police FDJP


Date post:	09-Dec-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Coaxial connector effects

Documents