Adv. Radio Sci., 9, 117–121, 2011
www.adv-radio-sci.net/9/117/2011/
doi:10.5194/ars-9-117-2011
© Author(s) 2011. CC Attribution 3.0 License.
Advances in
Radio Science
Two methods for transmission line simulation model creation based
on time domain measurements
D. Rinas and S. Frei
Dortmund University of Technology, Dortmund, Germany
Abstract. The emission from transmission lines plays an
important role in the electromagnetic compatibility of automotive electronic systems. In a frequency range below
200 MHz radiation from cables is often the dominant emission factor. In higher frequency ranges radiation from PCBs
and their housing becomes more relevant. Main sources for
this emission are the conducting traces. The established field
measurement methods according CISPR 25 for evaluation of
emissions suffer from the need to use large anechoic chambers. Furthermore measurement data can not be used for simulation model creation in order to compute the overall fields
radiated from a car. In this paper a method to determine the
far-fields and a simulation model of radiating transmission
lines, esp. cable bundles and conducting traces on planar
structures, is proposed. The method measures the electromagnetic near-field above the test object. Measurements are
done in time domain in order to get phase information and to
reduce measurement time. On the basis of near-field data
equivalent source identification can be done. Considering
correlations between sources along each conductive structure
in model creation process, the model accuracy increases and
computational costs can be reduced.
1
Introduction
Electromagnetic Compatibility plays an important role in the
development of automotive systems. Beside the radiating cable bundles integrated Electronic Control Units (ECUs) are
sources for electromagnetic emissions.
Standardized component field measurement methods, like
the ALSE antenna method provided in CISPR 25 for evaluation of electro-magnetic emissions from automotive sys-
tems, suffer from the need of large and expensive anechoic
chambers. Also a single field strength value is often not sufficient to characterize the EMI behavior of a complex system.
Furthermore it is not possible to use the measurement data
for behavioral simulation model creation. Having simulation models, a statement about the radiating electromagnetic
fields can already be made in early phases of development.
Basically the electro-magnetic emission must be distinguished in the emission of circuit boards and their housing
and the emission of their connecting cable bundles. Where
in lower frequency range up to 200 MHz radiation from the
bundles is dominant the importance of emission from ECUs
grows with the rising frequency.
In Rinas et al. (2010) a method for estimating the emissions from cable bundles is presented. Here the electromagnetic near-field at several points near a cable bundle is measured. With measured data a single transmission line model
with equivalent current distribution is generated. The measurements are done in Time Domain with a standard oscilloscope to get phase information and to reduce measurement
time.
To create a behavioral model of PCB the electromagnetic
near-field in a plane above the radiating structure is measured. Knowing the fields in an indefinitely extended plane
above the test object all information is available to calculate any field vector above this plane (Balanis et al., 1996).
This can be used to solve an inverse problem and to identify
the equivalent sources of the planar structure. This approach
is discussed in Vives-Gilabert (2007); Baudry et al. (2008);
Isernia et al. (1996); and Tong et al. (2010). Several methods dealing with the inverse problem are ill-posed due to the
presence of errors in measurement data, or suffer from the
problem of converging to local minima. When optimization
methods are applied this leads to long computation time (Isernia et al., 1996; Pierri et al., 1999).
Correspondence to: D. Rinas
(denis.rinas@tu-dortmund.de)
Published by Copernicus Publications on behalf of the URSI Landesausschuss in der Bundesrepublik Deutschland e.V.
ECU
(6)
λ
Plane
ECU
SGround
OURCE IDENTIFICATION OF CABLE BUNDLES
(Figure 2). Where Ri and RL are the terminating impedances
II.
asandshown
in exciting
Figure input
2. With
adherence
to (6)
phase jumps
V0 is the
voltage.
The number
of dipoles
is
As shown in [2] the current distribution of a radiating single
between
two
sources
from
∆φ
are
prevented
and
the
chosen
following
(3)different
.
This
approach,
in
comparison
with
the
Figure
1.
Simple
configuration
to
investigate
cable (Figure 1) can be estimated by measuring
the and
magnetic
118
D. Rinas
S. Frei:
Two
methods
for transmission
line simulation
model creation
method
of equally
distributed
dipoles,
can considerable
reduce
model
becomes
more
physically.
field in at least two field points along the cable.
the amount of necessary elementary sources, decrease
When calculation is based on two measurement sets
computational cost and increase the model accuracy.
E(t),H(t)
1
1
Dependencies between the neighboring elementary sources
Iz =
(Ve + ZI e )e jβ (l − z ) +
(Ve − ZI e )e − jβ ( l − z )
(1)
2Z
2Z
of each current path are used to correlate phases of the dipoles.
RL
For the spatial distribution
∆d the phase shift between
two
conductor
Ri
Cable
Ri
with
Ze
adjacent dipoles is set to
V0(t)
−1
2π
V0
Ve   a1 b1   I1 ( z1 ) 
∆ϕ =
∆d
(6)
 =

 ⋅
λ
(2)
Plane
ECU
 I  aECUb   I ( zGround

)
2
 e  2
 2 2 
as shown in Figure 2. With adherence to (6) phase jumps
structure
between two sources different from ∆φ are simple
prevented
and the
Figure
1. z.
Simple
Fig. 1. Simple
configuration
toconfiguration
investigate.
leads to current
Iz at position
Where
Ve andto investigate
Ie are voltage
model becomes more physically.
d current at the
endcalculation
of transmission
Z is the wave
When
is based on line,
two measurement
sets
pedance, β the propagation constant, l the length of the line,
sources
for emission
the conducting
1
1 of PCBs −are
and b includeAs
themain
and
I1 e and
Ipropagation
(Ve + ZI e functions
)e jβ (l − z ) +
(Ve − ZI
)e jβ ( lI− 2z ) are the (1)
z =
traces, the
distribution
of the 2approximating
sources can be
2Z
Z
rrent values corresponding to the magnetic field data.
Elementary
sources
R
L
conductor
bounded
Ri
with to the routing geometry. Furthermore techniques
The current
can
also be ofestimated
directly by
for distribution
correlating the
properties
the approximating
sources
∆φ
−1
V0
b1  model
Ve physically
 a1 field
 I1 ( z1 )can
along
creasing thetoamount
measured
points
the
cable
computeofa more
be
integrated
in the
 =


 ⋅
(2)
 respect
 the number of free
pending onmodel
the frequency
with
to reduce
creation process.
This
can
 I e  a 2 b2   I 2 ( z 2 ) 
source parameters and computation time. Furthermore the
simple structure model
leads toof
current
Whereand
Ve and
Ie are voltage
z at positionisz.given
possibility
error
accuracy
of
the
N
= l Icorrection
and current at the∆dend of transmission line, Z is (3)
the wave
Fig.
2. Drawin g of the test board (above), model of the board
results
increases.
2. Drawing of the test board (above), model of the board (below) with
impedance, β the propagation constant, l the length of the line,Figure(below)
with equally
(grey)
andthe
sources
Elementary
paper athe
for source
identification
transdistributed
sourcesdistributed
(grey)
and sources
sources only
along
currentonly
path
1method
a In
andthis
b include
propagation
functions
and I1 and of
I2 are
the equally
sources
along
the
current
path
(blue).
∆d
>
λ
(blue)
(4)
mission
lines,
in
form
of
cables
bundles
and
conducting
paths
current values corresponding
to the magnetic field data.
10
on PCBs, with considering correlations between the sources
The
current of
distribution
can also
be estimated
by To compute the dipole moments
∆φ
M the magnetic near-field
where N is
the
amount
measurement
points
and ∆ddirectly
the up
isincreasing
presented.
are done
in apoints
frequency
theInvestigations
amount of measured
field
along range
the cablehas to
where
N
is the amount
of
measurement
points
and 1d
the
be
measured
in
a
plane
above
the
planar
structure.
atial discretization
width
between
the
neighboring
points.
For
todepending
500 MHz.onThe
generatedwith
model
enables
the frequency
respect
to different types of
spatial
discretization
width
between
the
neighboring
points.
Afterwards the system of equations
hieving Iz the
discrete current
are approximated to a
postprocessing
e.g. farvalues
field estimations.
model
For achieving Iz the discrete current values are approximated
l
nusoidal function with nonlinear regression
algorithms.
N =
(3)
∆d
−nonlinear
1
to a sinusoidal function with
algorithms.
= Α(above),
H modelregression
Figure 2. Drawing of the testI 0board
of the board (below) with (7)
In order to2 create
an
equivalent
emission
model
elementary
In
order
to
create
an
equivalent
emission
model
elemenequally distributed sources (grey) and sources only along the current
path
Source identification of cable
bundles
1
∆d along
> λ the cable path in a (4)
ectric dipoles are placed in a line
(blue)in a line along the cable path
tary
electric
dipoles
are
placed
10
where Α contains the wave vector k0 and the fixed
(4). The of a
ight h over As
ground
the2010
cable,
in a height h over ground adopted from the cable, following
shownadopted
in Rinasfrom
et al.,
thefollowing
current distribution
geometric
parameters
for moments
each dipole
H the
complex
To compute
the dipole
M the and
magnetic
near-field
where
N
is
the
amount
of
measurement
points
and
∆d
the
pole moments
M
…M
are
calculated
with
3. The dipole moments M1 ...M K are calculated with
1
K
radiating
single
cablewidth
(Fig. between
1) can bethe
estimated
by measuring
has tofield
be measured
in a plane
above
the planar structure.
spatial discretization
neighboring
points. Formagnetic
at K near-field
points
is solved.
the
magnetic
field
in
two values
field points
along the cable.
AfterwardsIthe
system of equations
achieving
Iz the
current
are approximated
to a
Idiscrete
+ Iatk +least
1
k + Ik+1
M calculation
∆d k with
For
integrating
the correlation between sources into model
When
is based
on two
measurement
sets
sinusoidal
function
nonlinear
regression
algorithms.
k =
M
=
1d
(5)
(5)
k
2
2
I 0 = Α −1 H can be treated with
creation the inverse
problem
the
(7)
In1order to createjβ(l−z)
an equivalent
elementary
1 emission model
−jβ(l−z)
expandable
minimization
function
F
following
(7)
.
where Ik isIzelectric
the
z. a line
= discretized
(V
)eplacedI+
(Vealong
− ZIethe
)e cable path in
(1)a
where Ik is the discretized current Iz .
e + ZIare
ecurrent
dipoles
in
2Z
2Z
where Α contains the wave vector k0 and the fixed
height h over ground adopted from the cable, following (4). The
T
M
M dipole and H the complex
geometric
parameters −for
each

dipole
moments M1…M
calculated with
jϕ m
− jϕ m 
K are
with
III. SOURCE
IDENTIFICATION
OF
CONDUCTING
TRACES ON
F = magnetic
, y1 , at
z1 )KMnear-field
,...,points
α m is
( xsolved.
 ∑ α m ( x1field

∑
me
N , y N , z N )M m e
traces on planar
1
3m=1 Source identification ofm=conducting

(8)
TRUCTURES
PLANAR
S−1
= ∆d I k +I k +1
M
For
integrating
the
correlation
between
sources
into model
structures
k
(5)
T
Ve
a1 b1
I1 (z1 ) 2
),..., inverse
H 0 ( x N , yproblem
Minbe treated with the
0 ( x1 , y1 , z1the
N , z N )) →
=
·lines on planar structures, esp. (2) − (Hcreation
can
According to
Ie II transmission
a2 b2
I2 (z2 )
expandable
minimization
function
F
following
(7). planar strucwhere
I
is
the
discretized
current
I
.
k
z
According
transmissionparameters
lines on
CBs, can also be approximated
with elementary
dipoles. With
With
αm aretotheTable
fixed2 geometric
at observation
can also beand
approximated
T
owledge ofleads
the to
current
e.g. based
on CAD-data,
M
currentpaths,
Iz at position
z. Where
Ve and Ie arethe
voltagepointtures,
xn, yMnesp.
, zn,PCBs,
the amplitude
Mm is andwith
the− jϕelementary
phase
e-jφm of

− jϕ
III.
S
OURCE
IDENTIFICATION
OF
C
ONDUCTING
T
RACES
ON
F
=
α
(
x
,
y
,
z
)
M
e
,...,
α
(
x
,
y
,
z
)
M
e


dipoles.
of∑them current
e.g. based
∑ mWith
proximatinganddipoles
distributed
alongline,
these
m
N
N
N paths,
m
1
1 knowledge
1
current can
at thebeend
of transmission
Z ispaths
the wave
m =1
 m=1

(8)
PLANAR STRUCTURES
on CAD-data,
the approximating
dipoles
can
be
distributed
impedance, β the propagation constant, l the length of the
T
−
(
H
(
x
,
y
,
z
),...,
H
(
x
,
y
,
z
)
)
→
Min
0
1
1
1
0
N
N
N
to II transmission
lines onfunctions
planar structures,
esp.
along these paths (Fig. 2). Where Ri and RL are the termiline, According
a and b include
the propagation
and I1 and
PCBs, can also be approximated with elementary dipoles. With
nating
and Vgeometric
inputat voltage.
The
Withimpedances
αm are the fixed
parameters
observation
0 is the exciting
I2 are the current values corresponding to the magnetic field
knowledge of the current paths, e.g. based on CAD-data, the
point
xn, of
yn,dipoles
zn, the amplitude
and Mm is Eq.
and(3)
theThis
phase
e-jφ of
number
is chosen following
approach,
data.
approximating dipoles can be distributed along these paths
in comparison with the method of equally distributed dipoles,
The current distribution can also be estimated directly by
can considerable reduce the amount of necessary elementary
increasing the amount of measured field points along the casources, decrease computational cost and increase the model
ble depending on the frequency with respect to
accuracy.
N = l/1d
(3)
m
m
m
1d >
1
λ
10
Adv. Radio Sci., 9, 117–121, 2011
(4)
www.adv-radio-sci.net/9/117/2011/
Ri
magnetic field
h probe
V0
hs
V0
l
Magnitude [dBuA/m]
Magnitude [dBuA/m]
f0 = 4 MHz,
pulse/pause
th/tl =a 1,
and aplane.
risingItand
in thea height
h = 50ratio
mm over
ground
has falling
a length of
Figure 4 shows the magnetic field at point p1 in a frequency
edge of tlr,f==490
2.5mm
ns. and
Source
impedance
Ri1 ismm.
50-Ohm.
a thickness
of d =
It is terminated with a range of 10
1 MHz full
tofield500
MHz. The results in comparison with
simulation
Ze = 50 Ω impedance. The magnetic field is measured at two the full field
equivalent dipole
model with maximum error of about 3
0
simulation
agree
1) Single
Conductor
–
Measurement
Based
Results
positions
120Frei:
mm Two
and methods
y2 = 360for
mmtransmission
along the cable
at a dB. model creation
1 = S.
D. Rinas yand
line simulation
119
-10
The cable
3) consists
of a single conductor placed
height (Figure
hS = 20 mm
over ground.
in the height h = 50 mm over a ground plane. It has a length of
-20
10
magnetic
field probe with a
l = 490 mm and a thickness of d = 1 mm. It
is terminated
full field simulation
-30
Ze = 50 Ω impedance. The magnetic field is measured at two
equivalent dipole model
0
l
positions y1 = 120 mm andhs y2 = 360 mm along
the cable at a
-40
-10
height hS = 20 mm over ground.
Ze
-50
-20
1
10
100
500
Frequency [MHz]
-30
Figure 4. Magnetic field at p1 in comparison with full field simulation
-40
ground
3) Multiconductor – Simulation Based Results
-50
1
10
100
500 on
The
following investigations
are presently
based
Frequency [MHz]
computer simulations.
Ri
Figure 3. under
Singletest.
conductor under test Ze
Fig. 3. Single conductor
h
With the identified sources the magnetic field is calculated
The4. 4.
multiconductor
5) consists
offield
three
single
Fig.
Magnetic
fieldfield
at pat
in
with
fullfull
field
simulation.
1(Figure
Figure
Magnetic
p1comparison
in comparison
with
simulation
at a position p1 = (3, 3, 3)T [m]. Exemplary the fields of the
Dependencies between ththeth neighboring
conductors
placed
in
the
height
of
h
=
10
mm
over
ground
th
th elementary
ground
fundamental wave and the 2 , 7 , 15 and 20 harmonics of
plane.
length
conductor
is Model
l = 600
sources
of each
used to in
correlate
phases
Table
1. The
Magnetic
Field
ofeach
Measurement
based
at mm;
p1 in the
Multiconductor
–ofSimulation
Based
Results
the
pulsed
inputcurrent
signal path
are are
presented
TABLE
1. Asof a 3)
thickness is
d=
0.3
mmsimulation.
and they are arranged in a distance of
with
full
field
the dipoles.a full
For field
the spatial
distribution
1d the
phase
comparison
simulation
of the cable
under
test shift
given comparsion
The following investigations are presently based on
Figure
3.
Singledipoles
conductor
D = 5 mm. The internal source resistances are set to Ri = 50 Ω.
between
twosolver
adjacent
is under
set The
totest results agree with a
by
a MoM
[10] is shown.
computer
simulations. are Z = 50 Ω || jω·0.1nF, Z = 50 Ω,
The terminations
e1
e2
Magnitude
[dBµ A/m]
maximum error of 3 dB.
With the identified
sources
the
magnetic
field
is
calculated
2π
= 50 Ω +Frequency
jω·1µH.(Figure
TheSim.
excitation
is impressed
in the center
Model
TheZe3multiconductor
5) consists
of three
single
T
1ϕ
=
1d
(6)
at a position p1 λ= (3, 3, 3) [m]. Exemplary the fields of the
conductor.
conductors
placed in the height of h = 10 mm over ground
th
th
th
th
TABLE
I.
MAGNETIC
MEASUREMENT
MODEL AT
fundamental
wave
and
the 2 F,IELD
7 ,OF15
and 20 BASED
harmonics
ofP1 IN
COMPARSION WITH FULL FIELD SIMULATION
as shown
Fig. 2.areWith
adherenceintoTABLE
Eq. (6) phase
the pulsed
input in
signal
presented
1. Asjumps
a
between
different
from
1φ
are prevented
comparison
a full two
fieldsources
simulation
of Magnitude
the
cable
under
test givenand
[dBµA/m]
Frequency
the model
physically.
by a MoM
solver becomes
[10]
is more
shown.
The results agree with a
Sim.
Model
To
compute
maximum error of 3 dB.the dipole moments M the magnetic near-field
TABLE I.
4 MHz
29.4963
26.3542
plane. The length
of each
conductor
20 MHz
32.4243
32.9482is l = 600 mm; the
D mm and
thickness is d =600.3
they are37.3562
arranged in a distance of
MHz
36.7945
Z
Ri
D = 5 mm. The
internal
resistances
124
MHz source
43.1243
45.2324are set toe2 Ri = 50 Ω.
164 MHz
The terminations
are Z47.7698
|| jω·0.1nF, Ze2 = 50 Ω,
e1 = 50 lΩ50.1270
4 MHz
29.4963
has to be measured
in a plane
above the 26.3542
planar structure. Af- Ze3 = 50V0Ω + jω·1µH. The excitation is impressed in theZe1center
4 Results
conductor.
terwards the
system of equations
20 MHz
32.4243
32.9482
MAGNETIC FIELD OF MEASUREMENT BASED MODEL AT P1 IN
COMPARSION
WITH FULL FIELD SIMULATION
−1
I0 = A
60 MHz
H
124 MHz
36.7945
37.3562
h
(7)
Magnitude
43.1243 [dBµA/m]
45.2324
Frequency
where
A contains the wave vector k0 and the fixed geometric
50.1270
Sim.47.7698
parameters164
forMHz
each dipole
and H the Model
complex
magnetic field
at4KMHz
near-field points
is
solved.
29.4963
26.3542
For integrating the correlation between sources into model
20 MHz
32.9482
creation
the inverse 32.4243
problem can be
treated with the expandable
minimization
function
F
following
60 MHz
36.7945
37.35624.
124 MHz
43.1243
45.2324
M
X
F 164
= MHz αm (x1 ,y147.7698
,z1 )Mm e−j ϕm 50.1270
,...,
αm (xN ,yN ,zN )Mm e−j ϕm
m=1
(8)
With αm are the fixed geometric parameters at observation
point xn ,yn ,zn , the amplitude and Mm is and the phase
e−j φm of each dipole. H0 (xn ,yn ,zn ) stands for the measured
reference field at observation points. Optimization methods
are used for solving the minimization problem.
www.adv-radio-sci.net/9/117/2011/
D
Ze2ground
Ri
This chapter presents the results for model creation process
of a single radiating
and al smallunder
bundle.
Some reFigurecable
5. Multiconductor
test
V0 are presently derived from an electromagnetic full field
sults
Ze1
solver (EMCoS Consulting and Software, www.emcos.com).
h 3) is a pulsed signal
The input signal V0 (Figs. 1 and
Ze3
with an amplitude V0 = 5 V , a fundamental frequency of
f0 = 4 MHz, a pulse/pause ratio th /tl = 1, and a ground
rising
and falling edge of tr,f = 2.5 ns. Source impedance Ri is
50 Ohm.
4.1.1
!T
−(H0 (x1 ,y1 ,z1 ),...,H0 (xN ,yN ,zN ))T → Min
Ze3
Cable bundles
Figure 5. Multiconductor under test
m=1
M
X
4.1
Single conductor – measurement based results
The cable (Fig. 3) consists of a single conductor placed in
the height h = 50 mm over a ground plane. It has a length of
l = 490 mm and a thickness of d = 1 mm. It is terminated with
a Ze = 50  impedance. The magnetic field is measured at
two positions y1 = 120 mm and y2 = 360 mm along the cable
at a height hS = 20 mm over ground.
With the identified sources the magnetic field is calculated
at a position p1 = (3, 3, 3)T [m]. Exemplary the fields of the
fundamental wave and the 2th, 7th, 15th and 20th harmonics of the pulsed input signal are presented in Table 1. As
a comparison a full field simulation of the cable under test
given by a MoM solver (EMCoS Consulting and Software,
www.emcos.com) is shown. The results agree with a maximum error of 3 dB.
Adv. Radio Sci., 9, 117–121, 2011
d with a
d at two
ble at a
full field simulation
Magnitude [dBuA/m]
Radio disturbance characteristics – Limits and methods of measurement
As0 presented,
field approximation shows better
equivalentthe
dipole model
for the protection of on-board receivers”.
accuracy if the distribution of sources is matched with the
[2]
D.
Rinas, S. Niedzwiedz, S. Frei, “Far Field Estimations and Simulation
-10 distribution depending on the conductor path. The
current
In Figure
6 the magnetic field at point p in comparison with
Model Creation from Cable Bundle Scans”, EMC 1Europe, Wroclaw,
result
plots
show
also
a
smaller
deviation
if
the
phase
2010
full
field simulation
is shown.
The magnetic
agrees
120 -20
D. Rinas and S. Frei:the
Two
methods
for transmission
line simulation
model field
creation
correlation between sources is integrated in the model
[3] Constantine
A. Balanis,
Analysis
& Design”, Wiley,
with
a maximum
error“Antenna
of 3 dBTheory
at most
frequencies.
computation.
1996.
-30
10
Yolanda
Vives Gilabert, “Modélisation des emissions rayonées de
full field simulation
composants électroniques”,
Université de Rouen, 2007.
equivalent dipole model
0
[5] D. Baudry,
M. Kadi, Z. Riah, C. Arcambal, Y. Vives-Gilabert, A. Louis,
FULL
FIELD
SIMULATION
-50
B. Mazari, “Plane wave spectrum theory applied to near-field
1
10
100
500
Frequency [MHz]
measurements for electromagnetic compatibility investigations”, IET
-10
Science,
Measurement and Technology, 15. June 2008.
Full field
[6] Tommaso Isernia, Giovanni Leone, Rocco Pierri, “Radiation Pattern
simulation
Figure 4. Magnetic field at p1 in comparison with full field simulation
-20
Evaluation
from Near-Field Intensities on Planes”, IEEE Transaction on
und
Antennas and Propagation, Vol. 44, No. 5, May 1996.
3) Multiconductor – Simulation Based Results
[7] Xin
-30 Tong, D.W.P. Thomas, A. Nothofer, P. Sewell, C. Christopoulos,
Equally distributed sources Sources along current path
“A Genetic Algorithm Based Method for Modeling Equivalent Emission
The following investigations are presently based on
Sources of Printed Circuits from Near-Field Measurements”, APEMC
computer
simulations.
-40
Beijing,
2010.
No phase
1
10
100
500
lculated
correlation
[8] T. Isernia, G. Leone, R. Pierri, “Radiation
pattern evaluation from nearFrequency [MHz]
The
multiconductor
(Figure
5)
consists
of
three
single
of sources
field intensities on planes,” IEEE Trans. Antennas Propogat., vol. 44, pp.
s of the
conductors placed in the height of h = 10 mm over ground
701–710,
1996. field at p1 in comparison with full field simulation
Figure
6. May
Magnetic
onics of
Magnetic
field
at p1 in comparison
full field
simulation.
plane. The length of each conductor is l = 600 mm; the [9]Fig.
R. 6.
Pierri,
G. D’Elia,
F. Soldovieri,
“A two probeswith
scanning
phaseless
. As a
near-field far-field transformation technique,” IEEE Trans. Antennas
thickness
With phaseis d = 0.3 mm and they are arranged in a distance of
st given
In Figure 6correlation
the magnetic field at point p1 in comparison with
Propogat., vol. 47, pp. 792–802, May 1999.
D = 5 mm. The internal source resistances
are set to Ri = 50 Ω. B. Printed Circuit Boards – Simulation Based Results
with
[10] EMCoS Consulting and
www.emcos.com
FullSoftware,
field simulation
of sources
theafull field
simulation is shown. The magnetic field agrees
[4]
Magnitude [dBuA/m)
Table -40
2. Near-Fields of different Models in Comparison with full
TABLE
II.
NEAR-FIELDS OF DIFFERENT MODELS IN COMPARISON WITH
field
simulation.
e3
10
L AT P1 IN
conductor.
full field simulation
equivalent dipole model
Magnitude [dBuA/m)
0
D
Ze2
Ri
-10
l
V0
-20
Ze1
h
-30
-40
1
10
Ze3
100
ground
500
Frequency [MHz]
Figure 6. Magnetic fieldFigure
at p1 in
with under
full field
5. comparison
Multiconductor
test simulation
Fig. 5. Multiconductor under test.
B. Printed Circuit Boards – Simulation Based Results
4.1.2 planar
Single conductor
– simulation
resultsof a
A simple
structure is
analyzed. based
It consists
0.1 x 0.1 m plane and a conductor with total length of 0.16 m.
The following
are presentlyis based
comThe source
voltage investigations
is 1 Volt; termination
a 50onOhm
puter
simulations.
resistance. The magnetic field is measured in a height of
The ground
cable (Fig.
consists
of a single
conductor
0.01 m over
at 3)256
near-field
points.
The placed
sourcein
the
height
h
=
10
mm
over
the
ground
plane.
It
has
a length
parameters are computed with amplitude-data only. Simulated
of
l
=
600
mm
and
a
thickness
of
d
=
0.3
mm.
It
is
terminated
Annealing is used as optimization method for model parameter
with a serial circuit of resistor Re = 50  and inductance Le =
calculation.
1 µH. The model parameters are taken from the cable under
In TABLE
II source
near-field
calculations
models
test. The
internal
resistancefrom
is set different
to Ri = 50
. The
are shown
and compared
with a full
field
simulation
magnetic
field is measured
at two
positions
y1 = (MoM)
200 mmof
and
the planar
at along
a frequency
of at
100
MHz. Figure
7
shows
y2 structure
= 400 mm
the cable
a height
h
=
20
mm
over
S
the magnetic
field at point p1 = (3, 3, 3)T [m] in comparison
ground.
with the full
field4simulation
a frequency
of p
1 MHz
to
Figure
shows the inmagnetic
field range
at point
1 in a fre500 MHz.
quency range of 1 MHz to 500 MHz. The results in comwith the full
fieldapproximation
simulation agreeshows
with maximum
As parison
presented,
field
better
of about
3 dB. of sources is matched with the
accuracyerror
if the
distribution
current distribution depending on the conductor path. The
4.1.3 show
Multiconductor
– simulation
basedifresults
result plots
also a smaller
deviation
the phase
correlation between sources is integrated in the model
The following investigations are presently based on comcomputation.
puter simulations.
The multiconductor (Fig. 5) consists of three single conTABLE II.
-FIELDS OF DIFFERENT MODELS IN COMPARISON WITH
ductorsNEAR
placed
in the height of h = 10 mm over ground
FULL FIELD SIMULATION
A simple planar
structure
is analyzed.
It consists of a
Equally
distributed sources;
without phase correlation
Sources
along current path;with
without total
phase correlation
0.1 x 0.1 m plane and
a conductor
length of 0.16 m.
Sources along current path; with phase correlation
The source voltage
is 1 Volt; termination is a 50 Ohm
-40
resistance. The magnetic field is measured in a height of
0.01 m-60 over ground at 256 near-field points. The source
parameters are computed with amplitude-data only. Simulated
-80
Annealing is used as optimization method for model parameter
-100
calculation.
Magnetic Field [dBV/m]
The terminations are Z = 50 Ω || jω·0.1nF, Z = 50 Ω,
e1
e2
with a maximum
of 3 dB
at excitation
most
frequencies.
Z = 50 Ωerror
+ jω·1µH.
The
is impressed in the center
In-120
TABLE II near-field calculations from different models
are shown
and compared with a full field simulation (MoM) of
-140
the planar structure at a frequency of 100 MHz. Figure 7 shows
-160
[m] in
comparison
the magnetic
field at point
p1 = (3, 3, 3)T 100
1
10
200
500
[MHz]
with the full field simulationFrequency
in a frequency
range of 1 MHz to
500 MHz.
Figure 7. Magnetic field at p1 for different source distribution in comparison
Fig. 7. Magnetic field atwith
p full
for different
source distribution in comfieldapproximation
simulation
As presented, the 1 field
shows better
tra
and
can
nei
me
sim
[1]
parison with full field simulation.
accuracy if the distribution of sources is matched with the
current distribution depending
on the conductor path. The
V. CONCLUSION
result
plots
show
also
a
smaller
the the
phase
plane. The length of each conductordeviation
is l = 600ifmm;
In this paper
methods
for source
identification
of
radiating
correlation
between
sources
is
integrated
in
the
model
thickness islines
d = are
0.3presented.
mm and they
arranged
in a distransmission
The are
focus
is on cables
bundles
computation.
tance
of
D
=
5
mm.
The
internal
source
resistances
are set is
and conducting traces on planar structures. Model accuracy
to R
50. The by
terminations
are Zcorrelation
· 0.1 nF, the
e1 = 50||j ω
can
bei =increased
considering
between
Z
=
50,
Z
=
50
+
j
ω
·
1
µH.
The
excitation
is im-WITH
TABLE
II.
N
EAR
-F
IELDS
OF
DIFFERENT
M
ODELS
IN
C
OMPARISON
e2
e3
neighboring approximating sources.
FIELD SIMULATION
pressed in the centerFULL
conductor.
The
approaches
were
In Fig. 6 the magnetic fieldtested
at pointbyp1 simulations
in comparisonand
with first
measurements
and
were
compared
with
numerical
full
the full field simulation is shown. The magnetic field agreesfield
Full field data.
simulation
with a maximum error of 3 dB at most frequencies.
simulation
4.2
Equally distributed sources
[6]
[2]
[3]
[4]
[5]
[6]
Printed circuit boards
– simulation based results
REFERENCES
Sources along current path
simple 25
planar
is analyzed.
consists of
a –
[1]A “CISPR
Ed.3: structure
Vehicles, boats
and internalItcombustion
engines
0.1×0.1
m
plane
and
a
conductor
with
total
length
of
0.16
m.
Radio disturbance characteristics – Limits and methods of measurement
No phase
forsource
the protection
on-board
The
voltageof is
1 Volt;receivers”.
termination is a 50 Ohm resiscorrelation
[2]
D. Rinas,
Niedzwiedz,
“Far Field
Simulation
oftance.
sources
The S.
magnetic
fieldS.isFrei,
measured
inEstimations
a height ofand
0.01
m
Model
Creation
from
Cable Bundle
EMC Europe,
Wroclaw,
over
ground
at 256
near-field
points.Scans”,
The source
parameters
2010
are computed with amplitude-data only. Simulated Anneal[3]
Constantine
A. Balanis, “Antenna Theory Analysis & Design”, Wiley,
With
phase
ing1996.
is used as optimization method for model parameter calcorrelation
culation.
[4]
Yolanda Vives Gilabert, “Modélisation des emissions rayonées de
of sources
In
Table 2 near-field
calculations
different
composants
électroniques”,
Universitéfrom
de Rouen,
2007.models are
shown
and
compared
with
a
full
field
simulation
(MoM)A.ofLouis,
[5] D. Baudry, M. Kadi, Z. Riah, C. Arcambal, Y. Vives-Gilabert,
Adv. Radio Sci., 9, 117–121, 2011
Full field
simulation
Fig
B. Mazari, “Plane wave spectrum theory applied to near-field
measurements for electromagnetic compatibility investigations”, IET
Science, Measurement andwww.adv-radio-sci.net/9/117/2011/
Technology, 15. June 2008.
Tommaso Isernia, Giovanni Leone, Rocco Pierri, “Radiation Pattern
Evaluation from Near-Field Intensities on Planes”, IEEE Transaction on
Antennas and Propagation, Vol. 44, No. 5, May 1996.
[7]
[8]
[9]
[10
D. Rinas and S. Frei: Two methods for transmission line simulation model creation
the planar structure at a frequency of 100 MHz. Figure 7
shows the magnetic field at point p1 = (3,3,3)T [m] in comparison with the full field simulation in a frequency range of
1 MHz to 500 MHz.
As presented, the field approximation shows better accuracy if the distribution of sources is matched with the current distribution depending on the conductor path. The result
plots show also a smaller deviation if the phase correlation
between sources is integrated in the model computation.
5
Conclusions
In this paper methods for source identification of radiating
transmission lines are presented. The focus is on cables bundles and conducting traces on planar structures. Model accuracy is can be increased by considering correlation between
the neighboring approximating sources.
The approaches were tested by simulations and first measurements and were compared with numerical full field simulation data.
www.adv-radio-sci.net/9/117/2011/
121
References
CISPR 25 Ed.3: Vehicles, boats and internal combustion engines –
Radio disturbance characteristics – Limits and methods of measurement for the protection of on-board receivers, 2008.
Rinas, D., Niedzwiedz, S., and Frei, S.: Far Field Estimations and
Simulation Model Creation from Cable Bundle Scans, EMC Europe, Wroclaw, 203–208, 2010.
Balanis, C. A.: Antenna Theory Analysis & Design, Wiley, 1996.
Vives-Gilabert, Y.: Modélisation des emissions rayonées de composants électroniques, Université de Rouen, 2007.
Baudry, D., Kadi, M., Riah, Z., Arcambal, C., Vives-Gilabert, Y.,
Louis, A., and Mazari, B.: Plane wave spectrum theory applied to
near-field measurements for electromagnetic compatibility investigations”, IET Science, Measurement and Technology, 15 June
2008.
Isernia, T., Giovanni Leone, G. and Pierri, R.: Radiation Pattern
Evaluation from Near-Field Intensities on Planes, IEEE Transaction on Antennas and Propagation, 44(5), 701–710, 1996.
Tong, X., Thomas, D. W. P., Nothofer, A., Sewell, P., and
Christopoulos, C.: A Genetic Algorithm Based Method for
Modeling Equivalent Emission Sources of Printed Circuits from
Near-Field Measurements, APEMC Beijing, 2010.
Isernia, T., Leone, G., and Pierri, R.: Radiation pattern evaluation
from near-field intensities on planes, IEEE Trans. Antennas Propogat., 44, 701–710, 1996.
Pierri, R., D’Elia, G., and Soldovieri, F.: A two probes scanning
phaseless near-field far-field transformation technique, IEEE
Trans. Antennas Propogat., 47, 792–802, 1999.
EMCoS Consulting and Software, www.emcos.com, January 2011.
Adv. Radio Sci., 9, 117–121, 2011