SIST EN 50289-1-11:2017

SLOVENSKI STANDARD
01-februar-2017
1DGRPHãþD
SIST EN 50289-1-11:2002
.RPXQLNDFLMVNLNDEOL6SHFLILNDFLMH]DSUHVNXVQHPHWRGHGHO(OHNWULþQH
SUHVNXVQHPHWRGH.DUDNWHULVWLþQDLPSHGDQFDYKRGQDLPSHGDQFDSRYUDWQH
L]JXEH
Communication cables - Specifications for test methods - Part 1-11: Electrical test
methods - Characteristic impedance, input impedance, return loss
Kommunikationskabel - Spezifikationen für Prüfverfahren - Teil 1-11: Elektrische
Prüfverfahren - Wellenwiderstand, Eingangsimpedanz, Rückflußdämpfung
Câbles de communication - Spécifications des méthodes d'essai - Partie 1-11: Méthodes
d'essais électriques - Impédance caractéristique, impédance d'entrée, affaiblissement de
réflexion
Ta slovenski standard je istoveten z: EN 50289-1-11:2016
ICS:
33.120.20 äLFHLQVLPHWULþQLNDEOL Wires and symmetrical
cables
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

EUROPEAN STANDARD
EN 50289-1-11
NORME EUROPÉENNE
EUROPÄISCHE NORM
December 2016
ICS 33.120.20 Supersedes EN 50289-1-11:2001
English Version
Communication cables - Specifications for test methods - Part
1-11: Electrical test methods - Characteristic impedance, input
impedance, return loss
Câbles de communication - Spécifications des méthodes Kommunikationskabel - Spezifikationen für Prüfverfahren -
d'essai - Partie 1-11: Méthodes d'essais électriques - Teil 1-11: Elektrische Prüfverfahren - Wellenwiderstand,
Impédance caractéristique, impédance d'entrée, Eingangsimpedanz, Rückflußdämpfung
affaiblissement de réflexion
This European Standard was approved by CENELEC on 2016-09-05. CENELEC members are bound to comply with the CEN/CENELEC
Internal Regulations which stipulate the conditions for giving this European Standard the status of a national standard without any
alteration.
Up-to-date lists and bibliographical references concerning such national standards may be obtained on application to the CEN-CENELEC
Management Centre or to any CENELEC member.
This European Standard exists in three official versions (English, French, German). A version in any other language made by translation
under the responsibility of a CENELEC member into its own language and notified to the CEN-CENELEC Management Centre has the
same status as the official versions.
CENELEC members are the national electrotechnical committees of Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic,
Denmark, Estonia, Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia,
Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland,
Turkey and the United Kingdom.

European Committee for Electrotechnical Standardization
Comité Européen de Normalisation Electrotechnique
Europäisches Komitee für Elektrotechnische Normung
CEN-CENELEC Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2016 CENELEC All rights of exploitation in any form and by any means reserved worldwide for CENELEC Members.
Ref. No. EN 50289-1-11:2016 E
Contents Page
European foreword . 4
1 Scope . 5
2 Normative references . 5
3 Terms and definitions . 5
4 Test method for mean characteristic impedance (S type measurement) . 10
4.1 Principle . 10
4.2 Expression of test results . 10
5 Test method for input impedance and return loss (S type measurement) . 10
5.1 Method A: measurement of balanced cables using balun setup . 10
5.1.1 Test Equipment . 10
5.1.2 Test sample . 11
5.1.3 Calibration procedure . 11
5.1.4 Measuring procedure . 12
5.2 Method B: measurement of balanced cables using balun-less setup . 12
5.2.1 Test Equipment . 12
5.2.2 Test sample . 13
5.2.3 Calibration procedure . 13
5.2.4 Measuring procedure . 13
5.3 Method C: measurement of coaxial cables . 14
5.3.1 Test Equipment . 14
5.3.2 Test sample . 14
5.3.3 Calibration procedure . 14
5.3.4 Measuring procedure . 15
5.4 Expression of test results . 15
6 Test report . 17
Annex A (normative) Function fitting of input impedance . 18
A.1 General . 18
A.2 Polynomial function for function fitting of input impedance . 18
A.3 Fewer terms . 19
Annex B (normative) Correction procedures for the measurement results of return loss and
input impedance . 21
B.1 General . 21
B.2 Parasitic inductance corrected return loss (PRL) . 21
B.3 Gated return loss (GRL) . 23
B.4 Fitted return loss (FRL) . 25
B.5 Comparison of gated return loss (GRL) with fitted return loss (FRL) . 31
B.6 Influence of the correction technique on return loss peaks . 32
Annex C (normative) Termination loads for termination of conductor pairs . 35
C.1 General . 35
C.2 Verification of termination loads. 36
Bibliography . 37

European foreword
This document [EN 50289-1-11:2016] has been prepared by CLC/TC 46X "Communication cables".
The following dates are fixed:
• latest date by which this document has to be (dop) 2017-09-05
implemented at national level by publication of
an identical national standard or by
endorsement
• latest date by which the national standards (dow) 2019-09-05
conflicting with this document have to
be withdrawn
This document supersedes EN 50289-1-11:2001.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. CENELEC [and/or CEN] shall not be held responsible for identifying any or all such patent rights.
1 Scope
This part of EN 50289 details the test methods to determine characteristic impedance, input impedance and
return loss of cables used in analogue and digital communication systems.
It is to be read in conjunction with EN 50289-1-1, which contains essential provisions for its application.
2 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
EN 50289-1-1:2001, Communication cables - Specifications for test methods - Part 1-1: Electrical test
methods - General requirements
EN 50289-1-5:2001, Communication cables - Specifications for test methods - Part 1-5: Electrical test
methods - Capacitance
EN 50289-1-7:2001, Communication cables - Specifications for test methods - Part 1-7: Electrical test
methods - Velocity of propagation
EN 50290-1-2, Communication cables - Part 1-2: Definitions
3 Terms and definitions
For the purposes of this document, the terms and definitions given in EN 50290-1-2 and the following apply.
3.1
characteristic impedance
Z
C
(wave) impedance at the input of a homogeneous line of infinite length. The characteristic impedance Z of a
c
cable is defined as the quotient of a voltage and current wave which are propagating in the same direction,
either forwards or backwards.
uu
fr
(1)
Z
C
ii
fr
where
Z is characteristic impedance;
c
u is voltage wave propagating in forward respectively reverse direction;
f,r
i is current wave propagating in forward respectively reverse direction.
f,r
= =
3.2
mean characteristic impedance
Z
cm
in practice for real cables which always have structural variations the characteristic impedance is described
by the mean characteristic impedance which is derived from the measurement of the velocity of propagation
(EN 50289-1-7) and the mutual capacitance (EN 50289-1-5). However, this method is only applicable for
frequencies above 1 MHz and non-polar insulation materials (i.e. materials having a dielectric permittivity
which doesn’t change over frequency). The mean characteristic impedance approaches at sufficiently high
frequencies (≈100 MHz) an asymptotic value Z
∞
The characteristic impedance may be expressed as the propagation coefficient divided by the shunt
admittance. This relationship holds at any frequency.
aβ+ j β a
(2)
Zj≈−
c
jCω 1− j tanδωC ωC
( )
where
is complex characteristic impedance (Ω);
Z
c
α is attenuation coefficient (Np/m) ;
β is phase constant (rad/m);
tanδ is loss factor;
ω -1
is circular frequency (s );
C is mutual capacitance (F/m).
At high frequencies, where the imaginary component of impedance is small, and the real component and
magnitude are substantially the same we get for the mean characteristic impedance
τ
β 1
p
(3)
Z ≈==
cm
ω ××C C vC
Where
Z is mean characteristic impedance (m);
cm
v is velocity of propagation (m/s);
τ is phase delay (s/m);
p
C is mutual capacitance (F/m).
3.3
terminated input impedance
Z
in
impedance measured at the near end (input) when the far end is terminated by a load resistance of value
equal to the system nominal impedance Z
R
=
3.4
open/short input impedance
Z
OS
impedance measured at the near end (input) when the far end is terminated with its own impedance. In
practice this is the case when the round trip attenuation is greater than 40 dB at any measured frequency.
This property takes into account structural variations in the cable. For samples with lower round trip loss it is
determined by the open/short circuit method:
Z ZZ× (4)
os open short
where
is input Impedance of the cable obtained from an open/short measurement;
Z
os
Z is impedance with an open circuit at the far end of the cable;
open
Z is impedance with a short circuit at the far end of the cable.
short
3.5
fitted characteristic impedance
Z
fit
is obtained from a least square error function fitting of the open/short input impedance. The fitting can be
applied on the magnitude, real and imaginary part of the input impedance. The fitted characteristic
impedance is an alternative to the mean characteristic impedance to describe the characteristic impedance. It
is only valid if the variations with frequency of the input impedance around its characteristic impedance are
balanced.
3.6
(operational) return loss
RL
(operational) return loss is measured at the near end (input) when the far end is terminated by a load
resistance of value equal to the system nominal impedance Z . It quantifies the reflected signal caused by
R
impedance variations. The (operational) return loss takes into account the structural variations along the
cable length and the mismatch between the reference impedance and the (mean) characteristic impedance
of the cable (pair). If the (mean) characteristic impedance of the cable (pair) is different from the reference
impedance, one gets, especially at lower frequencies (where the round trip attenuation is low), multiple
reflections that are overlaid to the structural and junction reflections. Therefore, return loss RL is also
referenced as operational return loss.
As an example, Figure 1, shows the operational return loss under different conditions. The blue line shows
the return loss of a pair having a characteristic impedance equal to the reference impedance but taking into
account that the impedance is varying with frequency (see right-hand graph). The red line shows the return
loss of a pair having a characteristic impedance that is different from the reference impedance (110 Ω vs. 100
Ω). For both lines, periodic variations – that are caused by multiple reflections between the junctions at the
near and far end – are observed. The green line shows a simulation of a pair having a frequency independent
characteristic impedance which is equal to the reference impedance.
=
frequency dependent factor of the characteristic impedance
Return Loss
1,2 0,2
RLRL w/o mismatch
1,16 0,16
5 RLRL w mismatch
RLRL w /o mismatch; ZZcc frequency independent
1,12 0,12
1,08 0,08
1,04 0,04
1 0
25 Real
Imag
0,96 -0,04
0,92 -0,08
0,88 -0,12
0,84 -0,16
0,8 -0,2
0,1 1 10 100
0,1 1 10 100 MHz
MHz
Figure 1 — Return loss with and without junction reflections
3.7
open/short return loss
OSRL
way to avoid in the measurement of return loss multiple reflections due to a mismatch between the
characteristic impedance (asymptotic value at high frequencies) of the CUT and the reference impedance is
to use a CUT terminated in its nominal impedance and having a very long test length such that the round trip
attenuation of the CUT is at least 40 dB at the lowest frequency to be measured. For standard LAN cables,
this would result in a CUT length of roughly 1 000 m for the lowest frequency of 1 MHz.
Another way (when long CUT length is not available) is to measure the characteristic impedance (open/short
method) and to calculate the return loss. As the characteristic impedance is obtained from the measurement
of the open and short circuit impedance, it is proposed to name such obtained return loss open/short return
loss.
This open/short return loss includes the effect of structural variations and the mismatch at the near end
(including the effect due to a frequency-dependent characteristic impedance), but it does not take into
account multiple reflections.
Figure 2 shows the difference between operational return loss and open/short return loss. The left-hand
graph shows the results of a pair having a characteristic impedance which is different from the reference
impedance (110 Ω vs. 100 Ω). The right-hand graph shows the results of a pair having a characteristic
impedance which is equal to the reference impedance (100 Ω). One may recognize that the open/short return
loss does not take into account multiple reflections.
Return Loss Return Loss
0 0
RLRL w mismatch RLRL w /o mismatch
5 5
OSOSRLRL w/o mismatch
OSOSRLRL w mismatch
10 10
15 15
20 20
25 25
30 30
35 35
40 40
45 45
50 50
0,1 1 10 100 0,1 1 10 100
MHz MHz
Figure 2 — Return loss and open/short return loss
dB
dB
dB
3.8
structural return loss
SRL
The structural return loss is the return loss where only structural variations along the cable are taken into
account. The mismatch effects at the input and output of the transmission line (including the effect due to a
frequency-dependent characteristic impedance) have been eliminated. The structural return loss cannot be
measured directly but is calculated from the measurement of the characteristic impedance (open/short
method).
ZZ−
OS fit
SRL 20× lg (5)
ZZ+
fit
OS
where
Z is the (complex) input impedance obtained from the measurement of the open and short
OS
circuit impedance;
Z is the (complex) characteristic impedance obtained from a curve fitting of the real and
fit
imaginary part of ZOS.
The left-hand graph of Figure 3 shows the operational return loss, open/short return loss and structural return
loss of a CUT having a characteristic impedance of 110 Ω. A difference between both is observable. The
operational return loss takes into account all effects (structural variations, mismatch effects at the input and
output). The open/short return loss does not take into account mismatch effects at the output (i.e. no multiple
reflections). Whereas the structural return loss only takes into account structural variations along the cable.
The right-hand graph shows the real and imaginary part of the mean characteristic impedance (obtained from
the measurement of the open and short circuit impedance) and it’s fitting.
Return Loss Mean Characteristic Impedance
0 140 80
RLRL
OSOSRLRL 130 70
SSRLRL
120 60
110 50
100 40
Re(Zos)
fitted Re(Zos)
40 90 30
Im(Zos)
fitted Im(Zos)
80 20
70 10
60 0
50 -10
80 40 -20
0,1 1 10 100
0,1 1 10 100
MHz MHz
Figure 3 — Return loss, open short return loss and structural return loss
3.9
parasitic inductance corrected return loss
PRL
return loss where the effect a parasitic inductance (due to sample preparation and/or test fixture), which is
observed as an increase of the input impedance at high frequencies (above 100 Mhz), has been corrected
dB
Real Part [Ohm]
Imaginary Part [Ohm]
=
3.10
gated return loss
GRL
return loss where the effect of the test fixtures and sample preparation, which is observed as an increase of
the input impedance at high frequencies (above 100 MHz), has been corrected by a gating function
3.11
fitted return loss
FRL
return loss where the effect of the test fixtures and sample preparation, which is observed as an increase of
the input impedance at high frequencies (above 100 MHz), has been corrected by applying fitting function on
the input impedance
4 Test method for mean characteristic impedance (S type measurement)
4.1 Principle
This method shall only be applied for cables having non-polar insulation materials (e.g. PE, PTFE), i.e.
materials having a dielectric permittivity which doesn’t change over frequency. Or in other words this method
shall only be applied to cables having a mutual capacitance which doesn’t change over frequency.
The mean characteristic impedance shall be derived from the measurement of the velocity of propagation,
respectively phase delay, according to EN 50289-1-7 and the mutual capacitance according EN 50289-1-5.
The measurement shall be carried out at frequencies above 100 MHz where the phase delay approaches an
asymptotic value.
4.2 Expression of test results
The mean characteristic impedance Z shall be derived from Formula (6):
cm
τ
p
(6)
Z
cm
C vC×
where
Z is mean characteristic impedance (Ω);
cm
v is velocity of propagation (m/s), measured according EN 50289-1-7;
is phase delay (s/m), measured according EN 50289-1-7;
τ
p
C is mutual capacitance (F/m), measured according EN 50289-1-5.
5 Test method for input impedance and return loss (S type measurement)
5.1 Method A: measurement of balanced cables using balun setup
5.1.1 Test Equipment
The test equipment consists of a 2-port vector network analyser (VNA) with:
— S-parameter set-up;
= =
— Balun to convert the unbalanced signal of the VNA to a balanced signal. The balun shall have an
impedance on the primary (unbalanced) side equal to the nominal impedance of the measuring devices
(in general 50 Ω) and on the secondary (balanced) side equal to the nominal impedance of the CUT (e.g.
100 Ω) (the balun shall fulfil the requirements of Class A baluns as described in EN 50289-1-1);
— To perform a calibration of the test equipment (on the secondary side of the balun), a short circuit, an
open circuit and a reference load are required. The short circuit shall have negligible inductance and the
open circuit shall have negligible capacitance. The load resistor shall have a value close (within 1%) to
the nominal impedance of the CUT (e.g. 100 Ω) and with negligible inductance and capacitance;
— For the measurement of the input impedance and (operational) return loss a T-resistor network (see
Figure 4) is required to terminate the common and differential mode impedance at the far end of the
sample. The differential mode termination resistors shall be matched in pairs, each half the value of the
differential mode reference impedance Z (in general 100 Ω). If not specified otherwise, for example by
R
particular cabling standards, the common mode termination resistors shall be:
— 0 Ω for individually screened pair cables;
— 25 Ω for overall screened cables;
— 45 Ω to 50 Ω for unscreened cables.

Key
DM
R differential mode termination resistor
term
CM
R common mode termination resistor
term
Figure 4 — T-resistor network
5.1.2 Test sample
The CUT shall have or exceed the minimum length specified in the relevant sectional specification. Both ends
of the CUT shall be prepared, such that when connected to the terminals of the test equipment the influence
to the test result is minimised. The twisting of the pairs/quads shall be maintained.
5.1.3 Calibration procedure
It is not the intent of the standard to detail the algorithms applied by a VNA to correct the measured results
based on a calibration procedure but to detail the calibration procedure. Further information may be obtained
in the manuals of the VNA supplier.
The calibration shall be performed on the secondary side of the balun by applying consecutively an open,
short and load standard (see Figure 5).
Ri
open
reflection
Generator
bridge
short
Ri
Receiver
load
Figure 5 — Calibration set-up
5.1.4 Measuring procedure
The test sample shall be connected to the terminals of the test fixture. The scattering parameter S – i.e. the
xx
reflection coefficient Γ (in Formulae (9) to (13)) – shall be measured over the whole specified frequency range
and at the same frequency points as for the calibration procedure. The values shall be measured as complex
parameters. All pairs/quads shall be measured from both ends unless otherwise specified.
a) terminated input impedance (Z ) and (operational) return loss (RL)
in
Measure the scattering parameter S – i.e. the reflection coefficient Γ – of the CUT with the far end
xx
terminated by a load resistance as described in 4.1.1. Inactive pairs shall also be terminated by this T-
resistor network.
b) Open/short input impedance (Z ) and open/short return loss (OSRL)
OS
Measure consecutively the scattering parameter S – i.e. the reflection coefficient Γ – of the CUT with
xx
the far end in open and short circuit.
5.2 Method B: measurement of balanced cables using balun-less setup
5.2.1 Test Equipment
Method B is the preferred one for balanced cables for frequencies above 1 000 MHz as it avoids the use of
baluns which are often limited to 1 000 MHz. With this configuration it is possible to measure impedance and
return loss both of the differential and common mode.
Multiport vector network analyser VNA (having at least 4 ports) with
— S-parameter set-up;
— A mathematical conversion from unbalanced to balanced, i.e. the mixed mode set-up which is often
referred to as an unbalanced, modal decomposition or balun-less setup. This allows measurements of
balanced devices without use of an RF balun in the signal path. With such a test set-up, all balanced and
unbalanced parameters can be measured over the full frequency range;
— Coaxial cables – where the characteristic impedance shall be the same as the nominal impedance of the
VNA – are needed to interconnect the network analyser, switching matrix and the test fixture. The screen
of the coaxial cables shall have a low transfer impedance, i.e. double screen or more with a transfer
impedance less than 100 mΩ/m at 100 MHz. The screens of each cable shall be electrically bonded to a
common ground plane, with the screens of the cable bonded to each other at multiple points along their
length. To optimize the dynamic range, the total interconnecting cable attenuation shall not exceed 3 dB
at 1 000 MHz;
— To perform a calibration at the end of the coaxial interconnection cable coaxial reference standards, so
called calibration standards, i.e. a short circuit, an open circuit and a reference load, are required. An
alternative to the before mentioned open, short and load references is the use of an electronic multiport
calibration kit (E-cal module) which is supplied by the supplier of the VNA.
— If the calibration is performed at the test interface calibration reference artefact, i.e. a short circuit, an
open circuit and a reference load, are required. For further details refer to EN 50289-1-1.
— Termination loads as described in Annex C.
5.2.2 Test sample
The CUT shall have or exceed the minimum length specified in the relevant sectional specification. Both ends
of the CUT shall be prepared, such that when connected to the terminals of the test equipment the influence
to the test result is minimised. In case of balanced cables the twisting of the pairs/quads shall be maintained.
5.2.3 Calibration procedure
It is not the intent of the standard to detail the algorithms applied by a VNA to correct the measured results
based on a calibration procedure but to detail the calibration procedure. Further information may be obtained
in the manuals of the VNA supplier.
A full 4-port single ended (SE) calibration shall be performed. The calibration shall be either performed at the
ends of the coaxial interconnection cables or on the test interface.
In the first case open, short and load measurements (using coaxial reference standards, so called calibration
standards) shall be taken at the ends of the coaxial interconnection cables of each port concerned, and
through and isolation measurements shall be taken on every pair combination of those ports. One may also
use an electronic multiport calibration kit (E-cal module) which reduces significantly the calibration time. As
the calibration plane is the end of the coaxial interconnection cables the effect of the test interface is not
removed from the results which may have an influence in particular at high frequencies. To remove the effect
of the test interface it may be de-embedded from the measurement or a correction procedure applied. If the
effect of the test fixture is removed by de-embedding techniques it shall incorporate a fully populated 16 port
S-matrix. The de-embedded calibration shall not be performed by using only reflection terms (S , S , S , S )
11 22 33 44
or only near-end terms (S , S , S , S ).
11 21 12 22
When the calibration is performed at the test interface, open, short and load measurements (using calibration
reference artefacts) shall be taken on each SE port concerned, and through and isolation measurements
shall be taken on every pair combination of those ports.
5.2.4 Measuring procedure
The test sample shall be connected to the terminals of test fixture. The reflection coefficient Γ (in Formulae 9
to 13) – shall be measured over the whole specified frequency range and at the same frequency points as for
the calibration procedure. The values shall be measured as complex parameters. All pairs/quads shall be
measured from both ends unless otherwise specified. The reflection coefficient Γ (in Formulae (9) to (13))
corresponds to the scattering parameter S when measuring the differential mode impedance and S
DDxx CCxx
when measuring the common mode impedance.
a) terminated input impedance (Z ) and (operational) return loss (RL)
in
Measure the reflection coefficient Γ of the CUT with the far end terminated by termination loads as
described in Annex C.
b) Open/short input impedance (Z ) and open/short return loss (OSRL)
OS
Measure consecutively the reflection coefficient Γ of the CUT with the far end in open and short circuit.
The mixed-mode set-up allows to obtain all parameters in differential (denoted by the index DD) and common
mode (denoted by the index CC). Common multi-port VNA have integrated mathematical functions to convert
the single ended measurements to common mode or differential mode values (Formulae (7) and (8)).
S = (S − S − S + S )
DD11 11 21 12 22
(7)
S = (S + S + S + S )
CC11 11 21 12 22
(8)
where
S is scattering parameter S in differential mode;
DD11 11
S is scattering parameter S in common mode;
CC11 11
S is single ended scattering parameter.
xy
5.3 Method C: measurement of coaxial cables
5.3.1 Test Equipment
The test equipment consists of a 2-port vector network analyser (VNA) with:
— S-parameter set-up;
— Impedance matching adapter to convert the nominal impedance of the coaxial cable (e.g. 75Ω) to the
nominal impedance of the VNA (e.g. 50Ω), in case they are different;
— Open, short and load reference standards, traceable to an international reference standard.
5.3.2 Test sample
The CUT shall have or exceed the minimum length specified in the relevant sectional specification. Both ends
of the CUT shall be prepared, such that when connected to the terminals of the test equipment the influence
to the test result is minimised. In case of balanced cables the twisting of the pairs/quads shall be maintained.
5.3.3 Calibration procedure
It is not the intent of the standard to detail the algorithms applied by a VNA to correct the measured results
based on a calibration procedure but to detail the calibration procedure. Further information may be obtained
in the manuals of the VNA supplier.
If the nominal impedance of the CUT is equal to the nominal impedance of the VNA the calibration shall be
performed on the coaxial ports of the VNA, applying consecutively an open, short and load standard. If the
nominal impedance of the CUT is different from the nominal impedance of the CUT the calibration shall be
performed on the secondary side of the impedance matching adapter, by applying consecutively an open,
short and load standard.
5.3.4 Measuring procedure
The test sample shall be connected to the terminals of test fixture. The scattering parameter S – i.e. the
xx
reflection coefficient Γ (in Formulae (9) to (13)) – shall be measured over the whole specified frequency range
and at the same frequency points as for the calibration procedure. The values shall be measured as complex
parameters. The cable shall be measured from both ends unless otherwise specified.
a) terminated input impedance (Z ) and (operational) return loss (RL)
in
Measure the scattering parameter S – i.e. the reflection coefficient Γ – of the CUT with the far end
xx
terminated by a load resistance of a value equal to the system nominal impedance Z
R.
b) Open/short input impedance (Z ) and open/short return loss (OSRL)
OS
Measure consecutively the scattering parameter S – i.e. the reflection coefficient Γ – of the CUT with
xx
the far end in open and short circuit.
5.4 Expression of test results
All indicated parameters are complex parameters with magnitude and phase. Therefore all evaluations are
complex evaluations.
1+ Γ
( )
R
ZZ= (9)
in R
(1− Γ )
R
Z = ZZ× (10)
OS open short
1+ Γ open
( )
(11)
ZZopen =
R
1− Γ open
( )
1+ Γ short
( )
(12)
ZZshort =
R
1− Γ short
( )
RL = 20× lgΓ (13)
R
Z -Z
OS fit
SRL = 20× lg (14)
ZZ+
OS fit
ZZ-
OS R
OSRL = 20× lg (15)
ZZ+
OS R
KK K
12 3
ZK= + ++ (16)
fit 0
1/2 3/2
fff
The conventional reference impedance of the VNA for the differential mode is two times the nominal
impedance of the VNA, i.e. in general 2x50Ω=100Ω. The conventional reference impedance of the VNA for
the common mode is half the impedance of the VNA, i.e. in general 50Ω/2=25Ω.
If the specified reference impedance of the common or differential mode is different from the conventional
reference impedance of the VNA the return loss shall be corrected. In general a VNA has an integrated
function to change the reference impedance. Otherwise the return loss shall be corrected as follows.
*
ZZ−
*
in R
RL 20× lg (17)
*
ZZ+
in R
*
ZZ−
*
OS R
OSRL 20× lg (18)
*
ZZ+
OS R
where
Z is terminated input impedance in Ω when the far end is terminated by a load resistance
in
of value equal to the system nominal impedance Z
R;
Z is open/short input impedance in Ω;
OS
Z is (conventional) reference impedance in Ω;
R
* is specified reference impedance if different from the conventional reference
Z
R
impedance;
Z is input impedance with an open circuit at the far end in Ω;
open
Z is input impedance with a short circuit at the far end in Ω;
short
Z is fitted input impedance in Ω obtained from data fitting (see Annex A) of Z
fit in;
Γ is reflection coefficient measured with the far end terminated by a load resistance of
R
value equal to the reference impedance Z
R;
Γ is reflection coefficient with an open circuit at the far end;
open
Γ is reflection coefficient with an short circuit at the far end;
short
RL is (operational) return loss in dB for conventional reference impedance of the VNA;
* is (operational) return loss in dB for specified reference impedance if different from the
RL
conventional reference impedance of the VNA;
SRL is structural return loss in dB;
OSRL is open/short return loss in dB;
* is open/short return loss in dB for specified reference impedance if different from the
OSRL
conventional reference impedance of the VNA;
f is frequency in Hz;
K is fitting coefficients, see Annex A.
x
=
=
6 Test report
The mean characteristic impedance, input impedance and return loss shall be recorded as specified in the
relevant cable standard.
The test report shall include:
— temperature,
— frequency range,
— sample length,
— mean characteristic impedance, input impedance and return loss, as required.
Annex A
(normative)
Function fitting of input impedance
A.1 General
Function fitting of the impedance data is useful for separating structural effects from the characteristic
impedance when such effects are substantial. Where function fitting is used, the concept is that
measurements from nearby frequencies aid in the interpretation of the values obtained at a particular
frequency. Function fitting is based on a least error square function fit.
In general the measurement data may be fitted by a polynomial function:
y(x) = k ⋅ f (x) + k ⋅ f (x) + k ⋅ f (x) + . (A.1)
1 1 2 2 3 3
To obtain the fitting coefficients k a matrix is build in the form:
n
(A.2)
Y = A⋅ K
N N N N
   
 y ⋅ f (x )  f (x )⋅ f (x ) f (x )⋅ f (x ) f (x )⋅ f (x ) .
∑ i 1 i ∑ 1 i 1 i ∑ 1 i 2 i ∑ 1 i 3 i
   
i=1 i=1 i=1 i=1
k 
   
N N N N
 
   
k
y ⋅ f (x ) f (x )⋅ f (x ) f (x )⋅ f (x ) f (x )⋅ f (x ) .
∑ i 2 i ∑ 2 i 1 i ∑ 2 i 2 i ∑ 2 i 3 i  
  =   ⋅ (A.3)
 
i=1 i=1 i=1 i=1
k
   
N N N N
 
   
...
 
y ⋅ f (x ) f (x )⋅ f (x ) f (x )⋅ f (x ) f (x )⋅ f (x ) .
   
∑ i 3 i ∑ 3 i 1 i ∑ 3 i 2 i ∑ 3 i 3 i
   
i=1 i=1 i=1 i=1
 .   . . . .
   
The fitting coefficients k are then obtained from:
n
−1
K = A ⋅Y (A.4)
where
y(x) is polynomial function which fits the measurement data;
k is fitting coefficients;
n
is number of data points.
N
A.2 Polynomial function for function fitting of input impedance
Function fitting can be applied to the magnitude, real and imaginary components or phase of the input
impedance. The fitting shall be applied to open/short input impedance data (Z ). It works best with
OS
logarithmic spaced frequency data (most VNA offer this type of sweep) in that it results in appropriate
weighting of the lower and upper ends of a multi-decade frequency sweep.
The magnitude of the open/short input impedance it fitted by following polynomial:
KK K
12 3
(A.5)
ZK= + ++
fit 0
12 3 2
f
ff
The real and imaginary parts of the open/short input impedance are fitted by following polynomials:
KK K
12 3
(A.6)
Re(ZK)= + ++
fit 0
12 3 2
f
ff
KK K
12 3
(A.7)
Im(ZK)= + ++
fit 0
12 3 2
f
ff
The angle of the open/short input impedance is fitted by following polynomial:
KK K
12 3
(A.8)
∠()ZK+ ++
fit 0
12 3 2
f
ff
where
f is frequency in MHz;
K is fitting coefficients;
n
is fitting of the magnitude of the open/short input impedance;
IZ I
fit
Re(Z ) is fitting of the real part of the open/short input impedance;
fit
Im(Z ) is fitting of the imaginary of the open/short input impedance;
fit
∠(Z ) is fitting of the angle of the open/short input impedance.
fit
A.3 Fewer terms
Depending on the measurement frequency range and the amount of structural variation, usage of one or
more of the higher order terms may not be justifiable. The contributions from the higher order terms are
intended to be second order.
Where the data spans one decade or less, only the first two terms (or perhaps only the constant term) may
be justified. The resultant function fit is considered valid if it has a negative slope at low frequencies, is
asymptotic at higher frequencies and is free of oscillation with frequency. Two or three terms may be
sufficient when the data spans only one or two decades of frequency. If the capacitance is changing with
frequency as it does when polar dielectric material is present, more terms are generally justified.
Four criteria indicate use of fewer terms – check or have the computer program determine if the fitted meets
the following set of four criteria.
a) The fitted function, except when it is only a constant, has negative slope for frequencies below 3 MHz.
b) The 10 MHz fitted value is within the impedance range of +5 to –2 of the high frequency asymptote (fitted
constant value).
c) The area under the fitted function supplied by the frequency dependent terms on a log frequency basis,
exclusive of the constant area, is positive (constant component is not above the data).
=
d) The sum of the negative areas (those due to negative coefficients) is less than the total area due to the
frequency dependent terms.
If all four criteria are not met, the number of terms in the function shall be reduced by one by omitting the
highest order term. Otherwise, data spanning a wider range of frequencies and generally resulting in a better
fit must be obtained and fitted. The fit for impedance magnitude shall have a monotonic downward behaviour
with increasing frequency and approach a high frequency asymptote to a reasonable extent.
Annex B
(normative)
Correction procedures for the measurement results of return loss and
input impedance
B.1 General
The return loss of a transmission line (cable) can be obtained by different test methods, each having certain
advantages and/or disadvantages and therefore giving not exactly the same results. At higher frequencies,
the measured return loss is strongly influenced by the cable end preparation (stray inductances and
capacitances play an important role) leading to an underestimated cable performance. Under laboratory
conditions, one might be able to minimize this negative effect; however, under general industrial conditions
using automated test systems, one might need mathematical procedures to eliminate the effect of the cable
end preparation.
From the transmission line theory it is known that the characteristic impedance is decreasing with frequency
and approaching an asymptotic value (assuming dielectrics having an almost frequency-independent
dielectric permittivity). Therefore, an asymptotic value is assumed in the described correction procedures.
However, measurements at higher frequencies often show an increase of the characteristic impedance. This
is due to the cable end preparation and related stray inductances and capacitances. In fact we can consider
the CUT as a cascade of 3 transmission lines, the two cable ends and the cable, or even 5 lines taking into
account the test fixtures.
Figure B.1 shows the results of an S/FTP cable obtained with an automated test system. On the left-hand
side we see the inp
...