CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005

A PRACTICAL MAGNETIC FIELD EVALUATION METHOD FOR 10kV AIR-CORE REACTORS

Chunyan SONG*, Chunjie ZHU*, Shuwen DU†, Guoqi CHEN*, Zhengcai FU†
* Design Institute of Hangzhou Municipal Electric Power Bureau – P. R. China
† Dept. of Electrical Engineering, Shanghai Jiaotong University – P. R. China
zcfu@sjtu.edu.cn

INTRODUCTION µ 0 ρWI
Bz = [ Fz ( A1 , Z1 ) − Fz ( A1 , Z 2 )
2π ( z2 − z1 )(a2 − a1 )
In 110/10kV and 35/10kV substations in China, 10kV air-
core reactors are widely used as current-limiting reactors for − Fz ( A2 , Z1 ) + Fz ( A2 , Z 2 )] (3)
10kV compensating capacitor banks. But such reactors
produce significant magnetic fields in near areas. They are
Fρ = ∫ [R + cosθ ln( R + A − cosθ )]cosθdθ
π
important power frequency magnetic fields source in (4)
0
substations. Public concerns related to magnetic field
exposure necessitate the investigation of power frequency π  cosθ  R − Z 
magnetic fields in the vicinity of such reactors. Substation Fz = ∫  Z ln( R + A − cosθ ) + ln  
operation experiences also show that interference problems 0
 2  R+Z 
related to such reactors do occur in substation control rooms
near the reactors, especially to computer VDUs as often. The  Z ( A − cos θ )  
importance of predicting reactor magnetic fields at the − sin θ ⋅ tg −1   dθ (5)
substation design stage is then apparent.  R sin θ  

For evaluation of the magnetic flux density of air-core where, W is the number of coil turns, I is rms current,
reactors, many accurate or simplified calculation methods θ = ϕ '−ϕ , A = ρ ′ / ρ , A1 = a1 / ρ , A2 = a2 / ρ , Z = z ′ / ρ , Z1 = z1 / ρ ,
have been developed. But in these methods, detailed design Z 2 = z 2 / ρ , and R = (1 + A 2 + Z 2 − 2 A cos θ ) 2 . ρ ′ , ϕ ′ , z ′ is
1

parameters, such as the number of coils, width of air-gaps

between coils, conductor size and turns in each coil, current the coordinate of any point Q( ρ ′, ϕ ′, z ′) at the reactor cross
density, even insulation thickness etc., are required. Such section for the derivation. Other parameters are shown in
parameters are often the technical secret of reactor Fig.1.
manufacturers. It is difficult to the substation designers and
operators to get such parameters for various reactors. A z
practical magnetic field evaluation method, which does not
need such detailed reactor design parameters, is then
important to them. P( ρ ,ϕ , z )

A practical magnetic field evaluation method for air-core a a1 Q

reactor with sufficient accuracy for engineering application is a2
z2
proposed in this paper. It needs only the ‘nameplate data’ and
the outer size of the reactors to estimate the magnetic fields t
z1
around the reactor. ϕ y
o ϕ′

EXISTING METHOD FOR AIR-CORE REACTOR x

MAGNETIC FIELD CALCULATION
Fig.1 magnetic field model of a single-coil air-core reactor
For a single-coil air-core reactor, as shown in Fig.1, if the
details of the coil design are all known, at any point Because the area of interest is usually meters away from the
P( ρ , ϕ , z ) outside the reactor, the magnetic flux density can edge of the reactor, the magnetic field of 3-phase reactors can
be derived as, be calculated applying the principle of superposition,
neglecting the edge effect and vicinity effect at the ends of the
coil. Based on Eq.(1) to Eq.(5), a computational program is
B = Bρ2 + Bz2 (1) compiled to evaluate the magnetic field distribution in the
vicinity of the reactor. For example, with the detailed design
µ0 ρWI parameters, the radial magnetic field distribution of a
Bρ =  Fρ ( A1 , Z1 ) − Fρ ( A1 , Z 2 )
2π ( z2 − z1 )(a2 − a1 )  practical 10kV 3-phase air-core reactor shown in Fig.2 can
be calculated and shown in Fig.3 (B in µT, r in m, compared
− Fρ ( A2 , Z1 ) + Fρ ( A2 , Z 2 )  (2) with that of a single-phase reactor).

CIRED2005

Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005

(1) The cross-section size parameters are determined

according to the reactor outer size. In any case, the inner and
a4 outer diameter of the equivalent coil a1 and a2 are taken as
the inner and outer diameter of the total reactor (getting from
a3 product menu or nameplate, or even measured for reachable
a2 existing reactors).

a1 (2) The inductance L of the reactor can be calculated from the

given rated parameters, or read from the nameplate.
r
2b (3) The number of turns of the equivalent single-coil reactor,
Neq, is then calculated as

Neq = 4π L µ0 DaT (6)

(a) 3-phase reactor (b) 1-phase cross section
where, Da = a1 + a2 is the average diameter of the equivalent
1 a2 − a1 a2 − a1 2b 2b
560 coil, T = ， v = a + a = D . A,
，u = a + a = D
A + Bu + Cv 2 1 a 2 1 a

B, and C is taken from Tab.1 according to the range of u and

460 v value.
426
Tab.1 Determination of parameters A, B, and C
450 U V A B C
0～0.5 0~0.8 0.038 0.1204 0.118
426
0～0.5 0.8~1.6 0.036 0.1465 0.111
r
460 0.5~1.0 0~0.8 0.040 0.1065 0.136
0.5~1.0 0.8~1.6 0.019 0.1260 0.140
560
(4) The equivalent current density J is then calculated as,
(c) 3-phase model for calculation
NI
J= (7)
Fig.2 double-coil 3-phase reactors and its calculation model 2b( a2 − a1 )

1000 (5) The approximate magnetic flux density at any point

1- phase outside the reactor can then be calculated applying the
3- phase existing method that mentioned in the above section.
100

The magnetic field distribution of the above mentioned

10 double coil reactor, calculated by both the proposed practical
method and that with the detailed design parameters are
1 shown in Fig.4. Comparing the results, it is found that in
areas of 1m away from the reactor edge, the difference
between the results of the two methods is less then 3%.
0. 1
0 2 4 6 8 10
10000
Fig.3 magnetic field distribution of 3-phase reactors B / µT
exi st i ng
1000
pr oposed
100
THE PROPOSED PRACTICAL METHOD
10

The open parameters or nameplate data for an air-core 1

reactor include the rated voltage, rated capacity
(corresponding compensating capacitor bank capacity), rated 0. 1
reactance or inductance, rated current, and outer size 0 2 4 6 8 10
r/m
parameters. When the details of the design of the reactor is Fig.4 Magnetic field distribution evaluated by the proposed
unknown, a single-coil proximately equivalent reactor model and existing methods
can be established following the steps below.

CIRED2005

Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005

FIELD MEASUREMENT VERIFICATION OF THE Triangle layout takes a larger space but with lower magnetic
PROPOSED METHOD field level. At the orthogonal line of the horizontal layout
direction, the magnetic field level is lower than at the layout
line. These results could be referred in arranging the reactors
In 4 110/10kV substations in Hangzhou of eastern China, at the substation design stage.
magnetic field measurements are carried out near operating
10kV air-core reactors. The differences between the
measured results and the calculated data of both proposed
and existing methods are both under 10%, with some points
under 5%. The difference between calculated data of the two
method is only 3%. The bigger relative error between
measured and calculated data is because of the influence of
nearby current carrying conductors in the substations.

DISCUSSIONS

Impact of Reactor Size on Magnetic Field Levels (a) vertical layout (b) triangle layout

Applying the proposed method, compare two reactors with

the same rated electrical parameters but with different outer
size shown in Fig.5. Their magnetic field distributions are
shown in Fig.6. It is clearly shown that the one with a smaller
outer diameter has a lower magnetic field level.
(c) horizontal alyout
a2 a 2′
a1 Fig. 7 Vertical, triangle, and horizontal layout of 3-phase air-
a1′ core reactors
r
2b r
2b′ 1000
ver t i cal
B/μ T
hor i z- par a.
100
(a) small-diameter reactor (b) large-diameter reactor t r i angl e
Fig.5 Reactors with different outer size but same inductance hor i z- or t h.
10

1000 1
B/μ T small D
100 Large D 0. 1
r/ m
0 2 4 6 8 10
10 Fig.8 magnetic field distribution of 3-phase air-core reactors
under different layout
1

0.1 The Magnetic Field Interference Range of 10kV Air-

0 2 4 6 8 r/m 10 Core Rectors

Fig.6 Magnetic fields of reactors with different outer size

Impact and Optimisation of Reactor Layout
3-phase reactors are often arranged in horizontal, vertical and
triangle layout ways, as shown in Fig.7. Applying the
proposed practical method, the magnetic field distributions
under different layout ways are calculated as shown in Fig.8.
According to IEC 61000-4-8: 1993, Class I equipment (most
sensitive to power frequency magnetic field) is required to
have an immunity level of not less than 1.26µT. Considering
practical situations and compatibility margin, the critical
magnetic flux density of the interference domain is
determined as 1µT. Considering also the influence to areas in
upper and lower floors in the substation building, the
magnetic field distribution of reactor at 1.2m above the floors
(sensitive equipment often exist at this level) are shown in

CIRED2005

Session No 2
CIRED 18th International Conference on Electricity Distribution Turin, 6-9 June 2005

Fig.9. It is shown that tt 11m away from the reactor edge, the The impact of the layout of 3-phase reactor is also discussed.
magnetic flux density is decayed to less then 1µT. That Triangle layout can reduce magnetic field level, but needs
means for often seen reactors, the interference impact larger installation area.
distance is about 11~12m.
The interference range of reactor magnetic fields is also
analyzed quantitatively. The diameter of the interference area
1000 sel f - Fl . of often seen 10kV air-core reactor is less then 12m.
B/μ T upper - Fl .
100 l ower - Fl . These conclusions could be referred in substation design
practices.

10
REFERENCES
r/ m
1
0 2 4 6 8 10 12
[1] IEC, 1993, “IEC 61000-4-8:1993, Power frequency
Fig. 9 Magnetic field at 1.2m above different floors magnetic field immunity test”, Geneva, Switzerland.

[2] ICNIRP, 1998, “Guidelines for limiting exposure to time-

CONCLUSIONS varying electric, magnetic, and electromagnetic fields (up
to 300GHz) ”, Health Phys., vol.74, 494-522.

A practical magnetic field evaluation method for air-core
reactor is proposed. This method needs only the ‘nameplate
data” and the outer size of the reactors to estimate the
magnetic fields around it. The evaluation error of the
proposed method is less then 5% in areas 1m away from the
edge of the reactor, sufficient for engineering application.
Applying the proposed method, the impact on magnetic field
distribution of the outer size of the reactor design is also
investigated. The reactors with smaller outer diameter
produce relatively smaller magnetic fields.
[3] Q. Yu, S.A. Sebo, 1996, "Simplified magnetic field
modelling and calculation of large air-core reactor
coils", IEEE Trans. Magnetics. vol. 32, 4281-4283.
[4] Q. Yu, S.A. Sebo, 1998, "Accurate evaluation of the
magnetic field strength of large substation air-core
reactor coils ", IEEE Trans. Power Delivery. vol. 13,
1114-1119.

CIRED2005

Session No 2