Journal of ELECTRICAL ENGINEERING, VOL. 58, NO.

4, 2007, 207–213

ELECTROMAGNETIC FIELD CALCULATION

OF TRANSFORMER STATION 400/110 KV
ERNESTINOVO USING THE CDEGS SOFTWARE
∗ ∗ ∗∗
Srete Nikolovski — Predrag Marić — Zoran Baus

This paper presents calculations of the electric and magnetic fields of complex busbars, transformers and transmission
line structures at the transmission power station 400/110 kV Ernestinovo. By constructing and bringing the TS 400/110 kV
Ernestinovo in operation, it was expected that electric and magnetic fields would arise within and outside the fence of the
transformer station. The fundamental sources of low frequency electromagnetic fields in the TS Ernestinovo are the primary
elements of the 400 and 110 kV open substations as well as the connected 400 kV and 110 kV overhead lines. Due to
the complexity of the geometry of the substation elements, it was necessary to apply a three-dimensional approach to the
calculation and analysis of the electromagnetic field. Calculations of electromagnetic fields were conducted for quasi-static
states of electrical values at a power frequency of 50 Hz. The calculated electric and magnetic fields were compared with
permissible levels according to Croatian regulations. The maximum permissible levels for the electric field and magnetic flux
density were found to be exceeded at some locations in the station.
K e y w o r d s: electric field, magnetic field, transformer station, electromagnetic compatibility, moment method, com-
puter program

1 INTRODUCTION frequency of 50 Hz due to short-term effects of transients

which do not represent a risk in terms of electromagnetic
The book of regulations, which determines the per- influences.
missible levels of electric and magnetic fields, has been The law on non-ionizing radiation protection regu-
adopted on the basis of the Law on Non-ionizing radiation lates that appliances, substations and buildings which are
protection (article 10) which is in effect in the Republic of sources of electromagnetic fields or which contain sources
Croatia. The Law was passed and promulgated on Decem- of electromagnetic fields shall be used and brought in
ber 30, 2003. The regulations prescribe the obligations of operation only if they fulfil the requirements in compli-
utility owners to perform calculation or evaluation of the ance with their purpose, and when they do not expose
expected levels of electric and magnetic fields within and people to electromagnetic fields above the limiting levels
around their installations, ie, to provide a review on ful- prescribed by the regulator. The equipment within the
filment of the requirements prescribed by Articles 7, 8, 11 TS Ernestinovo generates a permanent 50 Hz electromag-
and 13 of the book of regulations. netic field. Consequently, we have the following limiting
This paper presents calculations of the electromag- levels of electric field intensity E and magnetic flux den-
netic field, both in the nearby environment and within sity B :
the transformer station itself and compares the results – for the field of professional exposure (8 hours per day)
with the permissible values prescribed by the Book of electric field intensity Eg8h = 5000 V/m and magnetic
Regulations. field density Bg8h = 100 µT,
New TS 400/110 kV Ernestinovo is in operation and – for the field of enhanced sensitivity (24 hours per day)
it was expected that electric and magnetic fields would electric field intensity Eg24h = 2000 V/m and mag-
arise within and outside the fence of the transformer sta- netic field density Bg8h = 40 µT.
tion. The fundamental sources of low frequency electro-
magnetic fields in the TS Ernestinovo are the primary
elements of the 400 and 110 kV open substations as well 2 MATHEMATICAL MODEL
as the connected 400 kV and 110 kV overhead lines. Due
to the complexity of the geometry of the above mentioned The calculations presented in this paper were carried
elements, it is necessary to apply a three-dimensional out with the HIFREQ [6, 7] program of the CDEGS [8]
approach to the calculation and analysis of the electro- software package. This program calculates the current
magnetic field. Calculations of electromagnetic fields were distribution in conductor networks consisting of rectilin-
conducted for quasi-static states of electrical values at the ear segments, and uses this current distribution to com-

∗
Faculty of Electrical Engineering, Power System Department Osijek, Croatia. E-mail: [email protected]
∗∗
SIEMENS d.d. Heinzelova 70 A, Zagreb, Croatia. E-mail: [email protected]

c 2007 FEI STU

ISSN 1335-3632
208 S. Nikolovski — P. Marić — Z. Baus: ELECTROMAGNETIC FIELD CALCULATION OF TRANSFORMER STATION . . .

pute the potential, electric field and magnetic field at se- be obtained by applying the superposition theorem as
lected points in space. The program can account for the an infinite sum of potentials caused by elementary time
presence of multiple horizontal layers of soil with different


variable charges dq̇ = λ̇ r ′ dl′ on the conductor surface.
electrical characteristics. The conductors and observation The infinite sum is treated as an integral for the elec-
points (ie, the points, where the potential and electro-


tric potential ϕ̇(r) caused by line charge density λ̇ r ′ ,
magnetic fields are computed) can be located in the air which is located at a point given by the position vector r ′
or in any of the earth layers. placed on all thin wire parts (structures), including the
The methodology used in the program is described in original conductors and their images with respect to the
detail in References 6 and 7. A brief summary is presented ground-air discontinuity, is given by the following integral
below. The potential and electromagnetic fields are first equation:
expressed in terms of the components of a vector poten-
tial, which is in turn expressed as a function of the cur-

λ̇ r ′ dl′


λ̇ r ′′ dl′′
1 1
Z Z
rents flowing in each segment of the conductor network. ϕ̇ = + (2)
4πε |r − r ′ | 4πε |r − r ′′ |
The currents in the conductor segments are determined l′ l′
by the requirement that the voltage drop between pairs
of points in the network be equal to the ZI drop along where:
a path connecting the points, so that in a complex repre-


λ̇ r ′ – phasor of line charge density of the original con-
sentation: ductor [A/m],
Z P2 Z P2

E · dl − ZIdl . (1) λ̇ r ′′ – phasor of line charge density of the mirrored con-
P1 P1 ductor [A/m],
Here, E is the complex vector of electric field intensity, |r − r ′ | – distance between the observation point and
Z is the internal impedance of the conductor and I is the the phasor of line charge density of the original conduc-
phasor of current flowing in it. Points and are normally tor [m],
taken as the mid-points of adjacent segments, and the |r − r ′′ | – distance between the observation point and
integrals are carried out over the shortest path linking the phasor of line charge density of the mirrored conduc-
them. tor [m],
This technique (which is a special application of the ε – permittivity of space in which the field is calculated,
method of moments) fully accounts for mutual interac- ie for air follows ε = ε0 ,
tions (conductive, capacitive, and inductive) between the l′ , l′′ – integration on the original (mirrored) conductor.
conductor segments, and also accounts for the potential Since the line charge density appearing under the in-
drop along the conductors due to their self-impedance. tegral sign is an unknown function, expression (1) has
Application of Equation (1) yields a matrix system to be solved using a numerical method due to the lack
that can be solved for the currents in the network. Once of a more appropriate analytical expression. In this pa-
the currents are known, they can be used to compute the per, the moment method (MOM), which has proven to
potential and the electromagnetic fields. be effective when solving electromagnetic tasks in unlim-
The computation of the fields due to a conductor seg- ited space, was used for solving Equation (1). In order
ment proceeds by integrating the contributions of electric to use the moment method for solving Equation (1), it
dipoles distributed along the axis of the conductor seg- is necessary to perform a discretization of this equation.
ments. The contribution of an individual dipole to the This is performed through the discretization of an un-
electromagnetic fields is evaluated with the help of a gen- known distribution of field source, ie, line charge density


eralized Sommerfeld integral that accounts for the pres- λ̇ r ′ by using a combination of appropriate number of N
ence of a multi-layered soil. linearly independent fundamental functions. Thereupon,
the discretization of conductor length on N segments and
A. Calculation of the electric field the discretization of observation points will be conducted.
First, it is necessary to divide the conductors into seg-
While calculating the conservative electric field, only ments of finite lengths ∆l (j = 1, N ), which can be a
conductors that are at known potentials are taken into line or ring depending on the desired accuracy. Then it
consideration. Furthermore, as the charges are located is necessary to approximate the unknown distribution of
at the surface of the conductors, it is necessary to dis- field source with the appropriate number of fundamental
tribute the charges appropriately on the conductor sur- functions λj in the form of the following expression:
face, which is later easily associated with the appropriate
mathematical tool and with the field equation. The con- N
X N
X
ductor surface was modelled with a thin one-dimensional λ̇ = a′j λ′j that is λ̇ = a′′j λ′′j (3)
wire grid. The effect of permittivity discontinuity (ε ) on j=1 j=1
the ground-air interface, ie, the effect of ground on the
electric conductor potential was taken into consideration where:
by using the mirror technique. The phasor of the elec- λ′j – fundamental function on segment j of the original
tric potential ϕ̇(r) located at some point in space can conductor,
 Journal of ELECTRICAL ENGINEERING VOL. 58, NO. 4, 2007 209

λ′′j – fundamental function on segment j of the mirrored Since each of the three components of the electric field
conductor. have a different sum and phase shift, the following can be
A constant on the segment was chosen, and a′j = 1 is applied:
on the observed segment, whereas it is zero on other seg- Ex (t) =Ex,max sin(ωt + ϕx )
ments. The same goes for a′′j . Taking into consideration Ey (t) =Ey,max sin(ωt + ϕy ) (7)
that the constant on the segment has been chosen as the Ez (t) =Ez,max sin(ωt + ϕz ) .
fundamental function and that great accuracy is required,
the number of segments per conductor has been increased where ω = 2πf and in our case f = 50 Hz. The absolute
so that the longest segment length does not exceed the value of the electric field intensity vector is defined by the
length of 1 m. Based on the above discussion, expression following expression
(1) can be written as: q
E(t) = Ex2 (t) + Ey2 (t) + Ez2 (t) (8)
N

1 X a′j λ′j r ′ dl′
Z


ϕ̇ r = + whereas the RMS value of the absolute value of electric
4πε j=1 |r − r ′ |
∆l′j field intensity is defined by the following expression:
N

1 X a′′j λ′′j r ′′ dl′′
Z
1 q 2
. (4) E= √ 2
Ex,max + Ey,max 2
+ Ez,max . (9)
4πε j=1 |r − r ′′ | 2
∆l′′
j

B. Calculation of the magnetic field

Since λ′ = −λ′′ , it is obvious that the expression (4)
has N unknowns on the right side. In order to solve the Space (air), where the rotational component of the
expression, N observation points in space where the po- EMF was calculated, has been modelled as a linear
tential is known, and which corresponds to the conduc- isotropic half-space having the same properties as a vac-
tors under voltage, have to be chosen. Thus the system uum, ie, ε = ε0 , µ = µ0 , κ = 0 . Discontinuity of these
of N equations with N unknowns is obtained, which is properties arises on the plane separating ground and air,
determined in matrix form as which is defined by the coordinate z = 0 . The electric
properties of the half-space below the limiting ground-
[ϕ] = [p][λ] (5) air plane are defined by the soil characteristics εr ≈ 80 ,
µ = µ0 and κ = 0.01 S/m. The coordinate system has
where the elements pi,j of the matrix system [p] repre- been chosen so that unit vector az is perpendicular to
sent the potential at observation point “i ” placed on the the limiting plane.
conductor surface of line current density λj . The magnetic field in the observed part of the half-
The Gauss-Seidel iterative method was used for solv- space is defined as:
ing the matrix equation. In this process, the iteration was Z ˙

considered successful the moment a relative accuracy of µ0 Idl × r − r ′
B= , (10)
10−7 was reached. Once equation (5) has been solved, 4π |r − r ′ |
lj
ie, an approximation of the line current density on the
conductors has been obtained, the vector-phasor of the where the electric field is defined by expression (6). Both
conservative component of electric field intensity on the expressions (6) and (10) are discretized, ie, the integral is
observation point with a position vector r can be deter- calculated as the sum of effects of all conductor segments
mined by using the following equation (6). at the observation point.
N


1 X a′j λ′j r r − r ′ dl′
Z


Ė r = +
4πε j=1 |r − r ′ |3
∆l′j

N


1 X a′′j λ′′j r r − r ′′ dl′′
Z
. (6)
4πε j=1 |r − r ′′ |3
∆l′′
j

Considering that a three-dimensional calculation of

the electric field intensity vector, which is caused by the
line charge density of a three-phase system that are mutu-
ally phase shifted by 120◦ and at the same time spatially
displaced (space between phases), is required, it should
be taken into account that electric field intensity is ellip-
tically polarized at every point in space. Fig. 1. Model of TS Ernestinovo used in calculations
210 S. Nikolovski — P. Marić — Z. Baus: ELECTROMAGNETIC FIELD CALCULATION OF TRANSFORMER STATION . . .

Fig. 2. Spatial distribution of electric field intensity inside TS Er- Fig. 3. Profile presentation of electric field
nestinovo

The electromagnetic field is determined at points lo- field. Consequently, the field of the balanced voltage sys-
cated at the nodes of a rectangular grid with equal mesh tem has been calculated since the unbalanced states (fail-
size at a height of 2 m above the ground, which lies in ures, switching operation, . . . ) are of short-term duration
the x- y plane, ie, parallel to the plane separating ground so that permanent system unbalance at these voltage lev-
from air. The space between two consecutive observation els can be disregarded.
points in the x and y directions is 2 m for observation
points that are in areas where the RMS value of the elec- C. Electric field intensity inside TS 400/110 kV
tromagnetic field is low and 1 m in areas where the RMS Ernestinovo
value is high.
The spatial distribution of the electric field inside TS
Ernestinovo is presented in Fig. 2 and Fig. 3. All calcu-
3 CALCULATION OF THE ELECTRIC FIELD lations have been conducted for the surface located 2 m
IN TRANSFORMER STATION 400/110 KV above the ground, since this is the highest field where
people could stay.
A graphical model of TS Ernestinovo used for the cal- The computed electric field much exceeds the permis-
culation of electric field is shown in Fig. 1, as modelled in sible limit inside the transformer station. The white line
the HIFREQ module of the CDEGS software [9]. While in Fig. 1 shows the maximum allowed field intensity of
calculating the conservative electric field, the phase volt- 5000 (V/m).
ages of a three-phase balanced system was imposed on
The greatest values occur under the 400 kV busbars
the conductors.
(14000 V/m), the connection conductors between 400 kV
The 760 elements were entered into the computer busbars and power transformers and on the 110 kV side of
model for the substation, along with each busbar, con- the substation under the 110 kV breakers and re-closers.
nection of voltage and current transformers, re-closers,
The calculated electric field for an eight-hour exposure
connectors with surge arresters, power transformers and
time for working personnel is colour coded and incorpo-
transmission lines.
rated into the drawing of the transformer station shown
The phase voltages of the conductors in the part of the in Fig. 4 and Fig. 5.
substation operated at 400 kV are:
The calculated electric field around the fence of the
transformer station is separated into four areas. In area I,
ϕA = 231214 ej0 V; ϕB = 231214 ej240 V; the field is less than 2000 (V/m), areas III and IV have
ϕC = 231214ej120 V. less than 1000 (V/m). The maximum is in the area II
under the 400 kV transmission lines (2000- 4000 V/m ),
which is greater than the limit (2000 V/m for 24-hours
The phase voltages of the conductors in the part of the
exposure time for humans).
substation operated at 110 kV are:

ϕA = 63584 ej0 V; ϕB = 63584 ej240 V; 4 CALCULATION OF THE MAGNETIC FIELD

IN TRANSFORMER STATION 400/110 KV
ϕC = 63584ej120 V.
Calculation of the magnetic field can not be accurately
The subject of this analysis is a long-term (eight-hour specified because of the impossibility to predict the cur-
or twenty four-hour) exposure of people to the electric rents flowing through the conductors at a given moment.
 Journal of ELECTRICAL ENGINEERING VOL. 58, NO. 4, 2007 211

Namely, loads are random incidents and unbalance are maximums occurring during exploitation and that were
unpredictable. Only severe seasonal and daily load vari- used for substation dimensioning, are specified. The prob-
ations can be predicted. Unbalance occurring at the 400 ability of such load occurrences is very low (and their du-
and 110 kV level can be disregarded so that the trans- ration is very short) since they represent emergency states
mission lines, power transformers and other parts of the that are not our primary concern. We are primarily in-
substation are considered to be loaded by a balanced sys- terested in stationary states. The loading currents should
tem of phase currents. be based both on the expected values of transmission line
and power transformer current ratings and power flow cal-
culation. Basic technical data for the transmission lines
and power transformers are selected for a temperature of
20 ◦C:
Loads for 110 kV transmission lines, in all cases Al/Fe
240/40 mm 2 and Iterm = 735 A, according to the data
from schematic R. Končar plan E02–400:

TL Ernest-Dakovo(1) , TL Ernest-Dakovo(2)
TL Ernest-Našice , TL Ernest-Osijek 1/1
TL Ernest-Osijek 1/2 , TL Ernest-Osijek 4
TL Ernest-Osijek 2/1 , TL Ernest-Osijek 2/2
TL Ernest-Vukovar , TL Ernest-Vinkovci

and loads for 400 kV transmission lines, in all cases

Al/Fe2x490/65 mm 2 and Iterm = 1920 A:

TL Ernes -Žerjavin, TL Ernest -Mladost

TL Ernest-Ugljevik
Fig. 4. Areas where an eight-hour stay is not allowed shaded
For the power transformers:
TR 400/110 kV, 300 MVA, primary current Ipr = 433 A,
secondary current Isek = 1506 A
For the purpose of estimating the load of the 400 kV
TL Ernestinovo-Žerjavinec, we can only use the data on
maximum loading from the year 1989 when this line was
connecting TS Ernestinovo and TS Tumbri with a value
over 1000 MVA, ie, 1445 A. Besides, the UCTE recom-
mends that the interconnection lines should not be loaded
more than 50 % of thermal line current, which in our case
is 960 A. A current of 1000 A/1000 A current will be used
for calculations of the magnetic field in the transformer
station.
Considering the above discussion, for the maximum
load, the following current values were used:
1. For 110 kV TL 70% of thermal line current,
2. For 400 kV TL 50% of thermal line current,
Fig. 5. Areas where an 24-hour stay is not allowed, shaded 3. For transformers maximum primary/secondary cur-
rents.
TS Ernestinovo has been connected to the transmis- The most unfavourable possibility of bus loading is to
sion network, which includes 13 transmission lines, 3 of connect all lines and transformers to one bus system. Con-
which are connected to the 400 kV network and 10 to sidering all loading possibilities, a distribution of loading
the 110 kV network. Two 400/110 kV power transform- currents through the 400/110 kV substation Ernestinovo
ers, both having a capacity of 300 MVA, have been in- is shown in Fig. 6.
stalled in TS Ernestinovo. Estimation of loading current Furthermore, it should be mentioned that such a high
should be based on an actual switching state and par- loading condition is not expected to occur inside a real
ticular substation elements. In the main project of TS substation but is used only as a worst case to calculate
Ernestinovo (L2, L3), loading currents that represent the the magnetic fields.
212 S. Nikolovski — P. Marić — Z. Baus: ELECTROMAGNETIC FIELD CALCULATION OF TRANSFORMER STATION . . .

sonnel for an 8-hour exposure time. This value is ex-

ceeded only in two places under the conductors between
the power transformers and the 400 kV and 110 kV bus-
bars.
The situation with the magnetic field density is much
better since only a small region exhibits values greater
than 100 µT as shown in Fig. 9.

Fig. 6. Model of loading currents

Fig. 8. Profile presentation of magnetic field density inside TS
Ernestinovo

D. Magnetic flux density inside TS Ernestinovo

Magnetic field computation was also performed using

the HIFREQ module of the CDEGS software. The spatial
distribution of the magnetic field inside TS Ernestinovo
is shown in Fig. 7. All calculations have been conducted
for an observation surface 2 m above the ground.

Fig. 9. Area where the magnetic flux density exceeds the permis-
sible level of 100 µT

Fig. 7. Spatial distribution of the magnetic field intensity inside

TS Ernestinovo 5 CONCLUSION

Figure 8 presents profiles of magnetic filed density High values of electromagnetic field can cause deleteri-
in µT. There were 140 profiles used for calculation of ous effects on humans exposed to them, so there is a need
magnetic field density on every 1 m of distance. and obligation according to Croatian law to calculate and
The permissible level for the magnetic field intensity measure these fields. This paper presents a mathematical
(B) in the transformer station is 100 µT for working per- model and calculations of electric field intensity (E) and
 Journal of ELECTRICAL ENGINEERING VOL. 58, NO. 4, 2007 213

magnetic field density (B) in a high voltage power trans- [7] DAWALIBI, F. P.—SELBY, A. : Electromagnetic Fields of
former station TS 400/110 kV Ernestinovo. Simulations Energized Conductors, IEEE Transactions on Power Delivery
8 No. 3 (July 1993), 1275–1284.
were performed using CDEGS software for an observation
[8] DAWALIBI, F. P.—DONOSO, F. : Integrated Analysis Soft-
plane at 2 m above the ground. The results are presented
ware for Grounding, EMF, and EMI, IEEE Computer Applica-
in 3D and 2D plots for the spatial distribution of electric tions in Power 6 No. 2 (April 1993), 19–24.
field (E) and magnetic flux density (B) in the station. [9] CDEGS User manual, SES technologies, Montreal, 2005.
Several areas in the substation were over the permissi-
ble values for the electric field intensity, especially under Received 12 December 2006
the 400 kV lines, busbars and around the power trans-
formers. For the magnetic field density, the situation is Srete Nikolovski DSc El Eng (1954) (M’1995, SM’2005)
not as bad, since only a small region under the 400 and was born in Belgrade on October 1, 1954. He obtained his BSc
110 kV busbars that connect power transformers exceed degree (1978) and MSc degree (1989), in electrical engineer-
the maximum allowed levels. ing from the Faculty of Electrical Engineering, University of
Belgrade and his PhD degree from the Faculty of Electrical
and Computing Engineering, University of Zagreb, Croatia in
Acknowledgement
1993. Currently he is a Full Professor in the Power System
The authors would like to thank Mr. Simon Fortin Department within the Faculty of Electrical Engineering, J.J.
from SES Technologies Ltd. Canada for his service and Strossmayer University in Osijek, Croatia. His main interests
constructive comments on the manuscript. are power system modelling, simulation and reliability. He has
published over 70 technical papers. He is a Senior Member of
IEEE Reliability Society, PES Society, EMC Society and a
References member of the Croatian National Committee of CIGRE.
Predrag Marić was born in Osijek on 11 December 1979.
[1] HAZNADAR, Z.—ŠTIH, Ž. : Electromagnetism 1 i 2, Školska He obtained a diploma degree in the field of Electrical Power
knjiga, Zagreb, 1997. Engineering from the Faculty of Electrical Engineering of Os-
[2] BARIČ, T.—NIKOLOVSKI, S. : Influence of Conductor Seg- ijek in 2004. His graduation thesis was awarded two prizes:
mentation in Grounding Resistance Calculation Using Boundary “Hrvoje Požar” of the Croatian Energy Institute as a spe-
Element method, Proceedings from PIERS (Progress in Electro- cially noted thesis in the field of power engineering, and a
magnetic Research Symposium), Pisa-Italy, 28.-31. 03. 2004. prize from the Croatian Power Company HEP for the best
[3] NIKOLOVSKI, S.—ŠLJIVAC, D.—JINXI, M. : Electromag- graduation thesis in the field of power engineering. Currently
netic Field Produced by Overhead Power Lines, Proceedings of he works as a research assistant in the Power System Depart-
the 8th International Conference on Operational Research KOI ment at the Faculty of Electrical Engineering, J.J. Strossmayer
2000 Hunjak, T., Scitovski, R. (ur.), Osijek : Zavod za matem-
University of Osijek. His main interest is modelling, simula-
atiku Sveučilišta u Osijeku, 2000.
tion and analysis of transient phenomena in power systems
[4] Technical documentation of TS 400/110 kV Ernestinovo project
and power system protection.
Rade Končar-Ingineering for energy and transport, d.d. Zagreb
2003. Zoran Baus, DSc El Eng (1951) was born in Pozarevac
[5] HAZNADAR, Z.—ŠTIH, Ž. : Electromagnetic Fields, Waves on June 15, 1951. He obtained his BSc degree (1975), MSc
and Numerical Methods, IOS Press, Ohmsha, Amsterdam, Vol- degree (1987), in electrical engineering and his PhD degree
ume 20, 2000. from the Faculty of Electrical and Computing Engineering,
[6] SELBY, A.—DAWALIBI, F. P. : Determination of Current Zagreb, Croatia in 2004. His major fields of interest include
Distribution in Energized Conductors for the Computation of intelligent control (fuzzy control and optimization) and power
Electromagnetic Fields, IEEE Transactions on Power Delivery system analysis. Presently, he is the Leader Project Manager
9 No. 2 (April 1994), 1069–1078. in Siemens dd. Croatia.

EXPORT - IMP ORT

of periodical s and of non-periodic ally
printed matters, book s and CD - ROM s

Krupinská 4 PO BOX 152, 852 99 Bratislava 5,Slovakia

s.r.o. s.r.o.
SLOVART G.T.G. GmbH tel.: ++ 421 2 638 39 472-3, fax.: ++ 421 2 63 839 485 SLOVART G.T.G. GmbH
e-mail: [email protected], http://www.slovart-gtg.sk
EXPORT - IMPORT EXPORT - IMPORT