EXPERIMENTAL TESTINGAND FEM ANALYSIS OF CORRUGATED TANKS
Mislav Ilijaevi, Stjepan Barun 1)
1)
Konar Distribution and Special Transformers Inc., Josipa Mokrovia 8, HR-10090 Zagreb, Croatia
Corresponding author: mislav.ilijasevic@koncar-dst.hr
1. Introduction
During the last decades of last century,
majority of manufacturers of distribution
transformers introduced corrugated tanks. The
main reasons were automated and cheaper
production and possibility of transformer
construction without conservator. Lack of
conservator means cheaper and more reliable
maintenance.
Ribs of corrugated walls and tank are constantly
exposed to stresses due to oil temperature
dilatation. Increase and decrease in oil volume
induces increase and decrease of pressure in
transformer tank. Proper design of ribs and tank
is crucial for safe and reliable operation.
Therefore, to properly design transformer tank,
it is vital that relation between oil volume
increase / decrease and pressure increase /
decrease is determined.
2. Experimental testing
In order to measure pressure change in
relation to change in oil volume, special test
equipment is designed and produced.
Equipment, together with one transformer
during testing is shown in Fig. 1. Test
equipment has capability to measure oil flow
(change in volume of oil), and pressure and
temperature in tank.
Fig. 1: Test equipment during testing on
transformer
During the test pressure levels varied between
-0.2 bar and 0.2 bar. These pressure levels are
proved not to induce stress levels high enough
to cause tank failure.
Experimental testing consists of pumping oil
into the tank up to the level at which point
pressure in the tank reached 0.2 bar, and
draining oil from tank up to the level at which
point pressure in the tank reached -0.2 bar, in
small increments. At each increment pressure
and oil volume are recorded.
Test was conducted on 5 transformers, ranging
in power from 50 kVA to 630 kVA.
3. FEM Analysis
FEM analysis provides very fast and
accurate way to calculate deformation (change
in volume) of ribs and tank for any pressure
level. 46 models of ribs were made, ranging in
rib depth from 100 mm to 300 mm, and ranging
in rib height from 500 mm to 1000 mm. Rib
dimensions are standardized, and designer
made a choice from standard dimensions during
designing. Fig. 2 shows one rib FEM model.
Fig. 2: Model of transformer rib
However, ribs are not the only construction part
of transformer that deforms under pressure.
Tank as a structure also deforms, that includes
tank bottom, cover and side walls. Therefore
transformer tanks were modelled, and models
Rib deformation p=0.15 bar p=0.25 bar
were made same as tanks used in experimental
testing to obtain best coincidence between Experimental 1.50 mm 2.45 mm
experimental and numerical testing.
FEM Analysis 1.55 mm 2.59 mm
FEM analysis showed that dependence between Tab. 1: Results comparison for 250 kVA
change in oil volume and pressure is linear transformer

V = xp (1) Some of transformers used for testing were also

used for endurance test according to EN 50464-
where x is tank elasticity coefficient and is 4 [1]. That made great opportunity to measure
calculated: p-V chart before and after endurance test. Fig. 4
4 shows results calculated by FEM analysis (red)
x = ni xi + xk (2) and experimental results (blue and yellow) for
i =1 100 kVA transformer.
where i - number of tank walls (i=1, 2, 3, 4), ni -
number of ribs on tank wall, xi - rib elasticity 8

coefficient, xk - tank structure elasticity 6

coefficient. 4

4. Comparative results experimental

V [l]

0 FEM analysis
-0,25 -0,2 -0,15 -0,1 -0,05 0 0,05 0,1 0,15 0,2 0,25 experimental after endurance test

-2

Fig. 3 shows results calculated by FEM -4

analysis (red) and experimental results (blue) -6

for 250 kVA transformer. -8

p [bar]

Fig. 4: Results comparison for experimental

12 testing and FEM analysis for 100 kVA
10 transformer
8

6
Endurance test induces high stress levels that
4

2
cause low cycle fatigue. Even if transformer
passes endurance test and low cycle fatigue do
experimental
V [l]

FEM analysis
0
-0,3 -0,2 -0,1 0 0,1 0,2 0,3
-2
not provoke transformer failure, its impact on
transformer is clearly visible on p-V chart.
-4

-6

-8

5. References
-10
p [bar]

Fig. 3: Results comparison for experimental [1] EN 50464-4 Three phase oil-immersed
testing and FEM analysis for 250 kVA distribution transformers - part 4:
transformer requirements and tests concerning
It is visible that there is a good compliance pressurized corrugated tanks, April 2007.
between experimental and numerical results.
Experimental results shows divergence from
linearity, and that is possible to explain by
residual stresses due to welding, hot-dip
galvanizing etc.
Maximum deformation of transformer rib is
also measured during experimental tests.
Comparison between measured deformation
and deformation calculated by FEM analysis is
shown in Tab. 1.