Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
Analysis of Stray Losses in Power Transformers by
3-D Magnetic Field Simulation
Chetan C. Adalja, M.L. Jain, Technology Department, EMCO Limited, Thane, India
Abstract—Transformer is a vital link in the power system,
which is connected in the network at different stages right from
the generating station to the user’s premises. The T&D losses
in the Indian power system ranges from 10-50%, which is
significantly high. In this, the contribution of transformers
exceeds 6% of the total power generated. Although the
transformer is the most energy efficient equipment in the
system, yet it would be expedient to make an attempt to further
reduce the losses in it to improve the overall system efficiency.
In this context, a case study was undertaken to analyze various
components of stray losses in power transformer and assess the
scope for their optimization.
The load losses in the transformer consist of I2R & stray
losses. In large rating transformer, the stray losses constitute
about 20-25% of the total load losses. Designers adopt various
cost-effective measures to minimize the losses and make the
transformer more efficient. These losses could be controlled to
a level of 8-10% by means of magnetic shunts judiciously
placed so as to canalize the leakage flux. However optimum
location of these shunts calls for accurate knowledge of 3-D
flux mapping. Some of the commercially available software
programs support 3-D field simulation studies for estimation of
stray losses with fairly good accuracy. Depending on the
accuracy requirement, these programs could be exploited to a
varying degree to achieve the desired results. This paper
presents a case study involving estimation of stray losses in a
100 MVA, 220/66/11 kV system transformer using an Integral
Equation Method (IEM) and Finite Element Method (FEM)
based EDMAG-3D software program. To present a
comprehensive picture of total stray losses, the winding stray
losses due to axial and radial magnetic fields are also calculated
using 2-D programs.
II. CASE STUDY
Accurate estimation of stray losses at design stage is a
prerequisite for a cost-effective and reliable design of
transformer. Towards this, a case study was carried out on a
100 MVA, 220/66/11 kV system transformer with S.C.
impedance at maximum, normal and minimum tap positions
of 10.46%, 10.20% and 10.04% respectively and load losses
of 245 kW (at normal tap) involving 3-D magnetic field
mapping and estimation of stray losses. As a first step, stray
losses in the transformer were estimated by to-the-scale
modeling of transformer and 3-D field mapping for a
standard design. Moreover, the influence of shunt
dimensions and edge stack construction in two halves on
stray losses was also studied. The solution to the problem
was attempted by plotting the 3-D magnetic field on both
HV & LV sides. Fig. 1 below shows the modeled HV side
3-D geometry using software program. Similarly, the LV
side geometry is modeled and analyzed.
I. INTRODUCTION
The stray losses in a transformer comprise winding stray
losses, viz. eddy loss and circulating current loss; the loss in
the edge stack (smallest packet of the core limb); and the
loss in structural parts, viz. frame, flitch plate and tank. Core
loss at the impedance voltage being insignificantly low, is
not considered in the present analysis. In case of large
generator transformers, stray losses due to high current
carrying leads also become significant. As the total stray
losses with shielding measures in large rating transformers
are of the order of 20-25% of the total load losses, it is
imperative to estimate stray losses accurately as control over
these gives a competitive advantage. Measures like using
judiciously designed magnetic shunts help reduce the stray
losses effectively [1].
The estimation of stray losses in structural parts of
transformer at design stage is generally carried out by using
empirical formulae covering wide range of design variants
and complicated asymmetrical geometries. These formulae
therefore inherently suffer form unpredictable inaccuracies,
which would be actually known only at final testing stage.
However, with the availability of high speed and accurate
computational tools and software programs [2] it is possible
to simulate complex geometries for 3-D electromagnetic
fields mapping and precise estimation of stray losses at
drawing-board stage.
Fig. 1. Modeled HV side geometry of Transformer
III. METHODOLOGY
The software tool, based on Integral Equation Method
(IEM) and Finite Element Method (FEM) is used for stray
losses analysis. This involves estimation of 3-D magnetic
field intensity (H, A/m) and induction (B, Tesla) together
with the eddy current losses in the structural parts and the
resultant temperature rises. It calculates values of the
magnetic field quantities at pre-defined locations in space,
as a sum of field created by the current sources (windings,
leads) with specified distribution of current using BiotSavart law and the field created by the fictitious magnetic
charges on the interface of magnetic and nonmagnetic media
(to account for ferromagnetic magnetization) using
algebraization of integral equations.
The complete transformer, comprising Core, Windings,
Frames, Flitch plates, Tank, Wall shunts and the epures
(pre-defined line on which magnetic field values are
computed in all 3-directions) is modeled for stray losses
estimation. The field quantities obtained at these epures are
used for estimation of stray losses.
498
Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
Table I & II show these values at normal and extreme tap
positions on HV and LV side tank respectively.
IV. ESTIMATION OF STRAY LOSSES
TABLE I
MAGNETIC FIELD STRENGTH ON HV SIDE TANK SURFACE
Magnetic field Intensity
Seff
Mode
(HV Side), A/m
(mm2)
Heff,
Heff_w
Hmax
Max. Tap
414
501
1601
48.63
Nor. Tap
607
743
2455
48.63
Min. Tap
644
795
2846
48.63
This section explains the modeling of transformer
geometries and estimation of stray losses in structural parts,
viz. tank, wall shunts, frames, flitch plates and edge stack.
The losses in different structural parts of transformer are
computed as follows.
A. Estimation of stray loss in Tank
The tank is made of mild steel having a nonlinear
permeability. The software tool first calculates the magnetic
field value at the tank surface by decoupling the effect of
nonlinearity. After estimation of the magnetic field, losses
are calculated considering nonlinearity by an iterative
estimation of coefficient of the tank influence factor.
The geometry of the tank is modeled by slicing it at
different heights and then connecting these levels using
vertically defined epures. Fig. 2 shows the local coordinate
system of the tank depicting the horizontal and vertical
epures.
TABLE II
MAGNETIC FIELD STRENGTH ON LV SIDE TANK SURFACE
Magnetic field Intensity
Seff
Mode
(LV Side), A/m
(mm2)
Heff,
Heff_w
Hmax
Max. Tap
422
478
1447
43.76
Nor. Tap
620
710
2337
43.76
Min. Tap
649
739
2453
43.76
The stray losses in tank computed from the above magnetic
field intensity values are presented in Table III below.
TABLE III
STRAY LOSSES IN TANK
Stray loss, kW
Mode
HV Side
LV Side
Max. Tap
6.54
7.06
Nor. Tap
6.48
7.00
Min. Tap
5.48
6.47
Total
13.60
13.48
11.95
The above trend of stray losses in tank follows the leakage
impedance pattern.
B. Estimation of stray loss in Wall Shunts
Shunts are made of CRGO material and modeled as
ferromagnetic bodies with linear permeability. The wall
shunts modeled on HV side of transformer tank are shown
in Fig. 1 above. The total 7 wall shunts are provided (viz. 3
on HV side, 3 on LV side and 1 on side wall) to reduce tank
stray losses. The epures are pre-defined at HV & LV side
wall shunts locations to estimate stray losses.
Fig. 4 below shows the plot of the modulus of flux density
components (Bx, By, Bz) along the height of the shunts
(opposite to the winding on LV side at central phase) for
normal tap position. The curves A, B & C indicate the
components of flux density, i.e. normal, along the width and
the height of the shunt respectively.
Fig. 2. Local co-ordinate system for modeling of tank
Fig. 3 below shows the variation of the modulus of flux
density components (Bx, By, Bz) along the height of the
tank surface (on HV side) opposite to the winding axis of
central phase and considering all wall shunts in place.
Fig. 3. Flux density variation on the HV Side tank surface at central phase
(opposite to winding axis)
In order to compute losses in the tank it is necessary to
obtain the values of Heff, Heff_w, Hmax and Seff. Where,
Heff: effective tangential magnetic field strength on the tank
surface (A/m); Heff_w: effective tangential magnetic field
strength (opposite to the winding axis) on the tank surface
(A/m); Hmax: maximum tangential magnetic field strength
on the tank surface (A/m) & Seff: loss emission area (mm2).
Fig. 4. Flux density profile in shunts placed on LV side of tank wall
For computation of stray loss in the shunts, it is necessary
to obtain values of B1, B2, L1 and L2 as represented by
notations in Fig. 4. The peak values of magnetic field at
normal tap position for first triangle is 0.01035T (B1) with
base 1025mm (L1) and second triangle is 0.01054T (B2)
with base 1645 mm (L2) respectively.
499
Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
The stray loss values in shunts, estimated based on above
magnetic field values, are indicated in Table IV below.
Mode
Max. Tap
Nor. Tap
Min. Tap
TABLE IV
STRAY LOSSES IN SHUNTS
Stray loss, kW
HV Side
LV Side
Side Shunt
0.78
2.51
0.54
1.38
2.54
0.36
0.76
1.47
0.25
typically at normal tap position, obtained for top and bottom
frame are as under. See Table V.
TABLE V
MAGNETIC FIELD CONCENTRATION IN FRAMES
Magnetic field (B), T
Top Frame
Bottom Frame
Maximum value (B1)
0.00744
0.02022
Minimum value (B2)
0.00032
0.00139
Total
3.83
4.28
2.48
It is observed that the stray loss values in HV side shunts
are lower than those on LV side due to their smaller height
and larger distance from the outer most winding.
C. Estimation of Stray Loss in Frames
Frames, also called yoke beams, are made of mild steel
material and are used for clamping of yokes and supporting
the windings. The frames are modeled as epures coinciding
with their physical locations for magnetic field plotting and
estimation of losses.
Fig. 5 & 6 below show the plots of the modulus of flux
density components (Bx, By, Bz) in top & bottom frames
along the height of the frame (from bottom to top) on the
HV side of the transformer at normal tap position. For
estimation of loss in the frames, it is essential to obtain the
maximum and minimum values of flux densities occurring
along the height of the frames, which is represented by
notations B1 & B2 in Fig. 5 & 6.
The field concentration in the bottom frame is over 2.7
times of that in top frame. This is attributed to lesser
distance between the winding bottom edge and the bottom
frame.
Similarly, the maximum and minimum field values are
obtained for top and bottom frames both for HV and LV
sides of transformer at extreme tap positions. The loss in the
frames calculated from magnetic field values is as shown in
Table VI below.
Mode
Max. Tap
Nor. Tap
Min. Tap
TABLE VI
STRAY LOSS IN FRAMES
Stray loss, kW
Top Frame
Bottom Frame
0.98
1.74
0.82
1.43
0.58
1.24
Total
2.72
2.25
1.82
The loss in the bottom frame, which is higher as compared
to the top frame, is commensurate with the higher flux
concentration in the bottom frame.
D. Estimation of Stray Loss in Flitch Plates
Flitch plates, made of MS and with slots at top and
bottom positions are used in the present case. The flitch
plates are 200 mm wide and 12mm thick modeled to the
scale, taking care of the slots and analysis carried out using
FEM technique. It is important to note that the stray losses
in such structural elements are quite low but the incident
magnetic field on them can be quite high for the exposed
area leading to unacceptable local hot spots. Fig. 7 & 8
shows the vector plot of eddy current density J (A/m2) and
temperature rise profile (K) from minimum to maximum
value differentiated by a colour band from blue to red, red
being the highest.
Fig. 5. Flux density variation along the height of the Top Frame
Fig. 7. Vector plot of current density J (A/m2) in Flitch Plate
Fig. 6. Flux density variation along the height of the Bottom Frame
It is observed that owing to the proximity effect, the
maximum flux density occurs in the bottom part of top
frame and the top part of bottom frame.
The maximum and minimum values of flux densities,
The magnetic field impinging on flitch plates induces
eddy currents. The eddy current loops are shown in both
solid and slotted regions in Fig. 7. The magnitude of normal
flux density being the highest at top and bottom winding
edges, it results in higher losses and hotspots in those
regions of the flitch plates. In order to avoid such situations,
the slots are provided in the flitch plates at both top and
bottom locations.
500
Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
normal tap position on the HV side of the transformer. The
curves A, B and C indicate the component of flux densities
normal to the edge stack, along the width of edge stack and
along the height of edge stack respectively.
As the magnitude of normal magnetic flux density is
higher at the top and bottom winding edges, Fig. 9
represents first and second triangle with peak value flux
densities 0.04270T & 0.04422T respectively (at winding
edges) along the height of edge stack. The length covered by
first and second triangle is represented by notations L1 & L2
in Fig. 9 is 635 & 710 mm respectively and distance
between the peaks of two triangles represented by notation
L12 in Fig. 9 is 1544 mm.
Based on the magnetic field and eddy current density, the
losses are calculated for principal and extreme tap positions
in the flitch plates as shown in Table VII below.
TABLE VII
STRAY LOSS IN THE FLITCH PLATES
Mode
Stray loss, kW
Max. Tap
0.65
Nor. Tap
0.62
Min. Tap
0.52
Fig. 8. Temperature profile in Flitch Plate
The temperature profile in the flitch plate is estimated by
specifying heat transfer co-efficient and using 3-D FEM.
Well, in absolute terms, the stray losses in flitch plates
may not form a significant part of the total losses of the
transformer [3]. Nevertheless, it deserves designer’s
attention as it could cause abnormal local hotspot rise in the
flitch plates, and that in-turn disintegration of oil in the close
vicinity, and consequential generation of fault gases, which
could be misconstrued as fault / defect in the transformer.
The effect of using non-magnetic material (stainless steel)
for flitch plate with following combination of slots was
studied and the results obtained are shown in Table VIII
below.
a) Flitch plate without slot
b) Flitch plate with slots at top and bottom
c) Flitch plate with slot(s) throughout winding height
Fig. 9. Flux density variation along the height of the Edge Stack
TABLE VIII
STRAY LOSS IN FLITCH PLATE WITH DIFFERENT DESIGNS
Stray loss, kW
MS Plate
SS Plate
SS Plate
SS Plate with
Mode
with slots at
without
with slots at
slot(s)
top &
slot
top &
throughout
bottom
bottom
winding height
Max. Tap
0.65
1.416
0.485
0.291
Nor. Tap
0.62
1.324
0.458
0.286
Min. Tap
0.52
1.248
0.425
0.252
From the above, it is observed that for a given design of
flitch plate,
a) Loss in SS plate without any slot is the highest
b) Loss in SS plate with slots at top and bottom is about
26% less than that with MS plate
c) Loss in SS plate with slot(s) throughout the winding
height is about 54% less than that with MS plate
E. Estimation of stray loss in Edge Stack
Stray loss in edge stack occurs due to flux impinging
normally (radially) on the outermost packet of the core.
For estimation of stay loss in edge stack it is essential to
compute the 3-D magnetic field values along & across the
height of the edge stack. Fig. 9 & 10 below show the plots
of the modulus of flux density components (Bx, By, Bz)
along and across the height of edge stack respectively, at
Fig. 10. Flux density variation across the height of the Edge Stack at top
winding edge position
The average value of magnetic field across the length of
edge stack is computed from Fig. 10. The maximum and
minimum value of magnetic field at top winding edge
position across the edge stack, represented by notations Bm1
& Bm2 in Fig. 10, is 0.07285T & 0.04379T respectively.
Similarly, the maximum and minimum value of magnetic
field at bottom winding edge is also obtained. These
magnetic field values are estimated for all phases at
principal & extreme tap positions on both HV & LV sides of
transformer. The stray loss based on above magnetic field
values estimated in edge stack is shown in Table IX below.
501
Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
respect to the windings, type and material. In the present
case study, the height of magnetic wall shunts was increased
by 645 mm on HV side of tank wall to attract larger chunk
of the leakage flux entering the tank and the results obtained
with above modification are shown in Table XII below.
TABLE IX
STRAY LOSS IN EDGE STACK
Mode
Stray loss, kW
Max. Tap
4.90
Nor. Tap
5.40
Min. Tap
4.38
F. Total stray load losses
The stray losses in winding i.e. eddy losses are also
measured as part of total stray losses during testing and are
practically inseparable; hence same are calculated through
another 2-D package and added to the structural losses to get
the total stray losses. The total stray losses in all structural
parts and windings are computed at normal and extreme tap
positions and the details are as summarized in Table X
below.
Sr.
No.
1
2
3
4
5
6
TABLE X
TOTAL STRAY LOAD LOSSES IN TRANSFORMER
Component
Stray losses, kW
Max. Tap
Nor. Tap Min. Tap
Tank
13.60
13.48
11.95
Shunts
3.83
4.28
2.48
Frames
2.72
2.25
1.82
Flitch Plates
0.65
0.62
0.52
Edge Stack
4.90
5.40
4.38
Winding eddy losses
27.67
27.03
21.86
Total Stray + Eddy losses
53.37
53.06
43.02
TABLE XII
COMPARISON OF ESTIMATED TANK LOSS WITH MODIFIED SHUNT
Reduction in
Tank stray loss, kW
Mode
loss (%)
Standard Shunt
Modified Shunt
Max. Tap
Nor. Tap
Min. Tap
13.60
13.48
11.95
12.13
11.68
10.59
It is observed that increase in shunt height results in
reduction in the tank loss significantly. This in turn does
have the effect of increasing the loss in the shunts, which is
marginal and hence ignored while reporting the total stray
losses with modification.
B. Modification in Edge Stack
In large transformer, the radially incident flux may cause
considerable eddy current loss in the edge stack, resulting in
abnormal local hot spots, thereby increasing the risk of
bubbling of oil in the local vicinity. Effect of division of the
edge stack on the stray loss was studied and the estimated
results are reported in Table XIII below.
Distribution of component stray losses, calculated as
percentage of the total stray load losses at normal tap
position is represented in Fig. 11 below.
Mode
Max. Tap
Nor. Tap
Min. Tap
TABLE XIII
COMPARISON OF LOSS IN EDGE STACK
Reduction in
Edge Stack stray loss, kW
loss (%)
Standard design
Modified design
4.90
2.18
55.51
5.40
2.57
52.42
4.38
2.09
52.27
The temperature profile of the edge stack is also analyzed.
The losses in the core blade packets including edge stack
and flitch plates are estimated and corresponding loss
density values entered into the program. The various heat
transfer co-efficients at outer core boundary surface are also
specified to solve planar temperature field in core blade
packets. Fig. 12 & 13 show the temperature profile of core
cross-section without & with division of the edge stack. The
temperature profile is differentiated from minimum to
maximum by blue to red colour band.
Fig. 11. Component stray losses as percentage of the total stray losses
The estimated values of stray losses are compared with
the tested values to validate the above results.
V. COMPARISON OF STRAY LOSS RESULTS
Comparison of the stray losses estimated by software
program and the measured test results is shown in Table XI
below.
Sr.
No.
1
2
10.80
13.33
11.41
TABLE XI
COMPARISON OF TOTAL STRAY LOSSES
Total Stray losses, kW
Component
Max. Tap
Nor. Tap
Min. Tap
Tested values
52.98
49.95
46.93
Estimated values
53.37
53.06
43.02
Deviation
-0.74 %
-6.22 %
8.33 %
Fig. 12. Temperature profile in standard edge stack design
The reference tested values vis-à-vis the estimated values
show a deviation of -0.74%, -6.22% & 8.33% at maximum,
normal and minimum tap positions respectively.
VI. CONTROL OF STRAY LOSSES
A. Shunt design modification
Magnetic shunts are effective in controlling the structural
stray losses as they offer high permeable path to the leakage
flux. The design of magnetic shunts depends on various
factors, viz. length, width and height, placement with
502
Fifteenth National Power Systems Conference (NPSC), IIT Bombay, December 2008
losses in the flitch plates.
VIII. FURTHER WORK
Precise estimation of stray losses is a subject in itself. It
may not be prudent to attempt very precise simulation for
computation of stray losses in routine designs disregarding
the economic considerations. However, application of
modern high speed and accurate computation tools offer
deep insight into the complex field phenomena in
asymmetric transformer geometries. There is a wide scope
to exploit these tools for development of new cost-effective
designs, exploring possibilities for improvements in certain
areas like shunt materials, use of yoke shunts, use of widthwise wall shunts [4] etc.
Fig. 13. Temperature profile in modified edge stack design
It is observed that in the present case the stray loss is
reduced by 52% at normal tap position and hotspot
temperature rise is reduced by 14 K after the division of
edge stack in two halves, which is quite significant.
ACKNOWLEDGEMENTS
The authors are grateful to the EMCO Management for
granting permission to publish this paper.
C. Total stray losses in transformer after modification
The total stray losses estimated in the transformer with
modified shunt and divided edge stack are presented in
Table IXV below.
REFERENCES
[1] Ramaswamy E, Sarma D V S, Lakhaini V K, “Design of magnetic and
non-magnetic shunts for a power transformer using EDMAG-3D”, XI
International Scientific Conference, Transformer Building-2005,
September 2005, pp. 70-77.
[2] Turowski, J., Turowski, M., and Kopec, M., “Method of threedimensional network solution of leakage field of three-phase
transformers”, IEEE Transactions on Magnetics, Vol. 26, No. 7,
September 1990, pp. 2911-2919.
[3] D A Koppikar, S V Kularni, PN Srinivas, S A Khaparde, R. Jain,
“Evaluation of flitch plate losses in power transformers”, IEEE
Transections on Power Delivery, Vol. 14, No. 3, July 1999.
[4] Prof.
S V Kulkarni & Prof. S. A. Khaparde, Transformer
Engineering – Design and Practice, Marcel Dekker, New York 2004,
pp. 169-230.
Sr.
No
1
2
3
4
5
6
TABLE IXV
TOTAL STRAY LOAD LOSSES WITH MODIFIED SHUNT
AND DIVIDED EDGE STACK
Stray losses, kW
Component
Max. Tap Nor. Tap Min. Tap
Tank
12.13
11.68
10.59
Shunts
3.83
4.36
1.35
Frames
2.72
2.25
1.82
Flitch Plate
0.65
0.54
0.62
Edge Stack
2.18
2.57
2.09
Winding eddy losses
27.67
27.03
21.86
Total Stray + Eddy losses
49.18
48.43
38.33
About the Authors:
Mr. Chetan C Adalja, born in
April 1982, a gold medalist from
Nirma University, completed his
graduation in Electrical Engineering
from CKPCET, Surat, South Gujarat
University in 2003, followed by postgraduation in 2005 in PAS-Power
Apparatus and Systems from Nirma
University, Ahmedabad. He started
his professional career as Lecturer at
D. Comparison of total stray losses after modification
The comparison of stray losses after modification in shunt
and edge stack is shown in Table XV below.
Sr
No.
1
2
TABLE XV
COMPARISON OF STRAY LOSSES AFTER MODIFICATION
Total stray losses, kW
Design
Max. Tap
Nor. Tap
Min. Tap
Standard
53.37
53.06
43.02
After Modification
49.18
48.43
38.33
Reduction
4.19
4.63
4.68
The results show that the modification in shunts and edge
stack effect reduction in the total stray losses by 4.19 kW,
4.63 kW & 4.68 kW at maximum, normal and minimum tap
positions respectively.
VII. CONCLUSIONS
1. Stray losses in a transformer can be precisely estimated
using EDMAG-3D software program that is a powerful
tool to aid fairly accurate 3-D field mapping of complex
transformer asymmetries.
2. The loss in the bottom frames is higher than the top
frames because of its close proximity with bottom edge of
the winding. It was observed that lowering of the bottom
frame height resulted in reduced frame losses. This is
attributed to its reduced interaction with the leakage field
returning to the bottom yoke.
3. The stray loss in edge stack is significant, leading to
localized hotspot. Division of the edge stack effects
substantial reduction in loss as well as the temperature.
4. Choosing appropriate material for flitch plate and
judicious slot dimensioning could effect reduction in stray
Engineering College in Surendranagar, Gujarat.He has been associated
with EMCO Limited from 2006 and working as a senior engineer in
Technology Department. He has authored 3 technical papers.
Mr. M.L. Jain, born in December
1945, completed his graduation in
Electrical Engineering from MNNIT,
Allahabad University in 1968,
followed by post-graduation in 1970
in
Design
and
Production
Engineering – Heavy Electrical
Equipment from MANIT Bhopal. He
started his professional career as
transformer design and development
engineer in BHEL Bhopal in 1971.
From 1979 onwards upto 1996, Mr. Jain was associated with testing
of transformers and other HV equipments. He has authored a chapter
on testing of transformers and reactors in BHEL monograph
‘Transformers’ published by Tata McGraw-Hill. Since 1996, Mr. Jain
has been associated with EMCO Limited. Having worked as Head of
Testing & Quality disciplines, he is presently Vice President –
Technology, responsible for up-gradation of transformer technology.
He has authored over 20 technical papers in the field of transformer
design analysis, testing and diagnostics. He is representing EMCO on
professional bodies like BIS and CBIP, and is a member of
CIGRE(I).
503