Mechanics and Mechanical Engineering

Vol. 20, No. 2 (2016) 109120

c Lodz University of Technology

Buckling Analysis of Cold Formed Silo Column

Karol Rejowski
Buildings Structures and Material Engineering
Gdansk University of Technology
[email protected]

Piotr Iwicki
Department of Structural Mechanics
Gdansk University of Technology
[email protected]

Received (20 January 2015)

Revised (26 January 2016)
Accepted (21 February 2016)

The paper is devoted to stability analysis of dierent models of steel cold formed silo
column. The steel cylindrical silos are often composed of corrugated walls and vertical
open-sectional columns uniformly placed along the silo circumference. Both the whole
3D silo, a simplied model consisting of one column with a part of the silo walls, and
a single column resting on elastic foundation provided by the silo walls were analyzed.
Linear buckling analyses were carried out using commercial FE package ABAQUS. Ax-
isymmetric and non-axisymmetric loads imposed by a bulk solid following Eurocode 1
were considered. The calculated buckling loads of 1D column model were compared
with the permissible one given by Eurocode 3 and with results found for the whole silo
and a single column on elastic foundation modeled by shell elements.
Keywords: buckling load, cylindrical orthotropic shell, columns, Eurocode approach.

1. Introduction
Silos can be built of thinwalled horizontally corrugated curved sheets strengthened
by vertical columns. The wall sheets carry circumferential tensile forces resulting
from horizontal wall pressure and vertical columns carry compressive forces exerted
by wall friction from a bulk solid. The silo column are therefore vulnerable to
buckling. For a small column distance the silo wall can be treated as an equivalent
orthotropic shell. The state of the art of the available recent knowledge on the
stability design of steel shells was summarized in [14]. The theory for eccentrically
110 Rejowski, K. and Iwicki, P.

stiened orthotropic cylinders was studied in paper [2]. Extensive measurements

of initial geometrical imperfections of stringerstiened cylinders were described in
[3]. The recent knowledge on thinwalled elastic shells with a homogenized regular
arrangement of rings and stringers was summarized in [4].
Currently some papers have dealt with the design of silos with corrugated walls
and sparsely distributed vertical stieners. Numerical linear and non-linear quasi
static analyses with initial geometric imperfections for a slender silo was performed
in [5]. The paper [6] dealt with failure of large cylindrical silos and proposed repair
methods based on sensitivity analyses. A bending theory for cylindrical orthotropic
shells with normal and shear pressures was presented in research [7]. The buckling
behavior of dierent corrugated wall silos with vertical stringers was analyzed with
a dynamic approach in [8].
In the design practice it is necessary to use a simple models instead of the anal-
ysis of the whole 3D silos. The Eurocode 3 [9] provides two approaches related
to global linear buckling of structure depending upon the silo column distance ds .
For a small column distance ds < ds,max the silo wall is treated as the equivalent
orthotropic shell (method A). For a higher column distance ds > ds,max , the
characteristic buckling load bearing capacity is based on the buckling formula for a
single column resting on the elastic foundation provided by the wall bending sti-
ness (method B). The 3D FE calculation results presented in [5, 8, 10] evidently
show that the Eurocode (2009) [9] approach may provide signicantly too conser-
vative outcomes for silos with corrugated sheets and columns. An improvement of
Eurocode 3 [9] buckling formula for the silo design basing on results of an approx-
imation of numerical nite element (FE) analyses was presented in research [10].
Dierent method based on application of formula for orthotropic shell theory with
a reduction factor within the range of method B was proposed in [8].
The aim of the present paper is to propose a simplify model of the silo composed
of horizontally corrugated sheets strengthened by vertical columns based on the
linear buckling analysis for a silo with sparsely distributer columns (within the
range of application of method B). In the present research a simplied model
consisting of one column with a part of the silo walls and a single column resting
on elastic foundation provided by the silo walls of stiness according to [11] are
compared with the buckling analysis of the whole silo and to existing Eurocode 3
formula. The buckling analysis of the silo with dierent number of columns allowed
to set a range of application of the proposed method.

2. Silo description
The numerical calculations were carried out with a real cylindrical metal silo be-
longing to a silo battery (Fig. 1). The height of a silo was H = 17.62 m and its
diameter D = 8.02 m (H/D = 2.2). The silo mantle consisted of 21 horizontally
corrugated sheets 890 mm 2940 mm 0.75 mm based on a foundation slab. The
silo was strengthened by 18 vertical columns composed of open thin-walled proles
with a varying cross-section in the form of the C (above 5m) and V (at height
0-5m ) - shape and thickness (t = 1.5-4 mm) along the column height uniformly
placed along the silo circumference at the constant distance of ds = 1.4 m. The
columns were connected to the wall sheets by screws.
 Buckling Analysis of Cold Formed Silo Column 111

Figure 1 Metal cylindrical silos with corrugated walls and 18 vertical columns at distance of 1.4 m
(height H = 17.62 m and diameter D = 8.02 m)

Figure 2 Vertical wall frictional traction and horizontal wall pressure in silo during emptying
according to Eurocode 1 [12] (pw vertical frictional traction, ph horizontal normal pressure)

The corrugation had 76 mm pitch and 18 mm depth. The silo roof was made from
metal sheets inclined under an angle of 25o to the horizontal and stiened by 36
radial beams.
The silo wall loads induced by maize were calculated according to Eurocode
1 [12]. During axisymmetric emptying, the standard maximum wall normal and
shear stress in the bin were ph = 30.52 kPa and pw = 19.38 kPa, respectively. When
112 Rejowski, K. and Iwicki, P.

considering a possible non-symmetric emptying, they increased up to ph = 36.26 kPa

and pw = 26.68 kPa (Fig. 2). In numerical analyses it was assumed that the load
level according Eurocode 1 [12] was the reference value, i.e. the load factor was
always related to the wall shear stress of pw = 26.68 kPa.
In our previous numerical analysis [8] the shell nonlinear static and dynamic
analysis with geometrical and material nonlinearity of the above described silo was
presented. The silo shell model consisted of 884880 S4R elements [8]. Than the
model of single silo column resting on elastic foundation of dierent stiness was
again investigated in [11, 13] (Fig. 3). The results of the analysis [11, 13] revealed
that for the foundation stiness described by Eurocode 3 [9] and according formula
proposed in [11] the global buckling occurred.

Figure 3 Evolution of column buckling load factor against lateral foundation stiness k [11, 13]

Present research is a continuation of work [8, 11, 13]. The above described silo was
again analyzed with variable number of columns on the silo perimeter from 3 to 50.
Due to the fact that the silo shell model required huge amount of nite elements
the silo simplied model was introduced. The silos walls were modeled as socalled
equivalent orthotropic shell with the stiness according to [9] and the silo columns
were modeled with 2node beam elements (B33). Such model of dierent silo was
veried with the results for the shell 3D model in [10]. The analysis was performed
by commercial program ABAQUS [14]. The element size in the model was 0.07 m
0.07 m for the silo wall and 0.07 m for the beam elements in silo columns. The
number of the S4 elements in silo wall was 90720. Number of the B33 elements in a
single silo columns was 4563. Total number of the nite elements in whole silo was
91476 for silo with three columns and 95256 for silo with fty columns (Fig. 4a).
 Buckling Analysis of Cold Formed Silo Column 113

a)

b) c)

Figure 4 FE mesh for the silo model with wall as equivalent orthotropic shell and columns
modeled with 2node beam elements: a) whole silo model, b) simplied silo model and c) boundary
conditions (BC) for the side edges of the part of the silo model (restrained displacement directions)

The steel was assumed to be elastic (linear buckling analysis) with the following
properties: modulus of elasticity E = 210 GPa, Poissons ratio = 0.3. In the
present analysis such as in Eurocode 3 [9] normal wall pressure is neglected. The
horizontal wall loads would be transferred to the silo corrugated curved sheets and
result in circumferential tensile forces.
The investigation was focused on the research of the model of a part of the
silo that would have similar buckling load as the whole structure within a range of
sparely distributed columns. A section of cylinder with a single column was analyzed
(Fig. 4b). The silos walls were modeled as equivalent orthotropic shell and the silo
columns were modeled with 2node beam elements (B33). The element size in the
model of the part of the silo was similar to above described model of the whole
structure. The total number of nite elements in simplied model of the silo with
eighteen columns was 15372 (about 6 times less than for the whole silo with walls
modeled by equivalent orthotropic shell and columns modeled with beam elements).
114 Rejowski, K. and Iwicki, P.

Than a single silo column resting on elastic foundation provided by the silo walls
according to Eurocode 3 [9] formula and to the proposed stiness [11] for dierent
distance of columns for 1D beam model of column were performed.

3. Eurocode approach and the proposed modification

According to Eurocode 3 (2009), there are two methods for calculating the stability
of a cylindrical silo with vertical columns. The applicability of these methods is
dependent on the spacing of the columns ds . The rst method named in this paper
as the method A is based on an orthotropic shell theory. It is used for spacing
column ds < ds,max , according Eqs (3) and (4) with satisfying results in comparison
with 3D FEM analysis [8, 10]. The second method named in this article as method
B is based on the solution beams on elastic foundations (Fig. 5). For the case
of the column distance (ds > ds,max ). The characteristic buckling load bearing
capacity N of single vertical column in a metal silo with corrugated walls is given
by the following formula in Eurocode 3 [9]

N = 2 EJ K (1)

with
D
K = ks and D = 0.13Etd3 (2)
d3s
where EJ is the bending stiness of columns in the plane perpendicular to the
wall, the foundation stiness K denotes the bending stiness of corrugated sheets
between vertical columns, ks = 6 is the coecient, D denotes the wall sheet bending
stiness, E is the modulus of elasticity, t denote the sheet thickness and d is the
sheet height. The following assumptions were met to lay down Eqs (1) and (2): the
number of buckling halfwaves along the circumference is equal to the half of the
columns number, the column is loaded by vertical forces prescribed at both ends
only (horizontal pressure is not considered). However, in the Eq. (2) the silo wall
curvature is not taken into account.
For the small column distance ds < ds,max , where [10]
( )0,25
r 2 Dy
ds,max = kdx , (3)
Cy

there exists the approach in Eurocode 3 [9] for the silo buckling based on an or-
thotropic shell theory. The critical buckling resultant force nx,Rcr per the unit
circumference of an orthotropic shell should be evaluated at each appropriate silo
level by minimizing Eq. (4) with respect to the critical circumference wave number
j and the buckling height li as
( ( ))
1 A2
nx,Rcr = min 2 2 A1 + , (4)
j A3

where: Dy the exural rigidity parallel to the corrugation, Cy the stretch-

ing stiness parallel to the corrugation and r the cylinder radius and kdx the
coecient recommended to be taken as 7.4, j the circumference wave number,
 Buckling Analysis of Cold Formed Silo Column 115

the parameter including buckling height li and A1,2,3 the parameters includ-
ing the exural and stretching stiness in orthogonal directions of the equivalent
orthotropic shell.

a) b)

Figure 5 Buckling of silo column (Eurocode 3 [9]) (a) vertical beam supported by elastic founda-
tion and loaded by vertical forces at both ends, (b) stiness determination of wall sheets during
column buckling

The load bearing capacity N of a single vertical column in metal cylindrical silos
should be apparently always smaller than the plastic force [9]

Nb,Rk = Aef f fy , (5)

where Aef f is the eective crosssectional area of the column and fy denotes the
yield stress.
Due to the column distance in the silo of Fig. 1 ds = 1.4m > ds,max = 1.16
m, the appropriate standard formula for the silo buckling strength is described by
Eqs (1) and (2). Using Eqs (1) and (2), the characteristic buckling strength was
exceeded in 1 column prole twice for symmetric emptying and in 2 proles about
2.5 times for non-symmetric emptying [8]. The column buckled for the characteristic
wall shear stress equal to pw = 10.67 kPa (the limit load factor = 0.4 according
to Eqs (1) and (2)). In turn, the buckling strength calculated by Eqs. (3) and
(4) was almost 7 times higher than this by Eqs. (1) and (2). This discontinuity
of the Eurocode 3 [9] buckling capacity in function of the number of silo columns
was described in [8, 10] (Fig. 6). As the method A gives satisfying results in
comparison with 3D FEM analysis [8, 10] the discontinuity of relation between
the silo buckling capacity and the number of columns (Fig. 7) may be caused by
inaccuracy of the method B.
In method B the circumferential wall curvature in the bending stiness of
corrugated sheets was not included (Fig. 5b). Modication of the stiness determi-
nation of the wall sheets during column buckling was determined in [11] with the
static scheme assumed in Fig. 6. In the proposed formula [11] the circumferential
wall curvature, and a component associated with compression was included.
116 Rejowski, K. and Iwicki, P.

Figure 6 Static scheme for modication of stiness determination of wall sheets during column
buckling

Figure 7 Buckling load factor of silo versus number of vertical columns for the silo walls mod-
eled with equivalent orthotropic shell and columns modeled with beam elements: a) whole silo,
b) a part of the silo (Fig. 4b), and according to Eurocode 3 [9]: c) method B with proposed
K [11], d) method B e) method A

4. Silo buckling capacity

The computed buckling load factor (LBA) for a variable number of columns around
the perimeter of the silo for the whole silo model (Fig. 4a) and for a section of the
silo with a single column (Fig. 4b) by taking into account only a shear stress in
the wall are presented in Fig. 7. Results of numerical analysis were compared with
the procedures according to Eurocode 3 [9] for variable number of columns around
 Buckling Analysis of Cold Formed Silo Column 117

the perimeter (method A and B) and to buckling load factor for single column
resting on elastic foundation according to the proposed in [11] formula for elastic
stiness. The calculated rst buckling eigenmodes are given in Figs. 89.

4.1. Calculation of whole silo model

The buckling load factor for the silo with a variable number of columns increased
with the number of column (Fig. 7). The number of circumferential halfwaves
was depended on the number of columns (Fig. 8a). Based on the linear buckling
analysis, three ranges of the column spacing were determined, which corresponded
to three dierent buckling forms (Fig. 7). In the rst range of the column spacing
(Fig. 8) the silo buckling mode had three halfwaves for every two sections between
the silo columns in circumferential direction, so every second column was subjected
to exural buckling. Other columns were twisted. The buckling mode of the silo
in the rst range of the column spacing corresponded to the arc deformation with
boundary conditions according to the model form Fig.6. Buckling mode within the
second range (12 to 22 columns around the circumference, or the column spacing
between 1.15 to 2.1 m) was characterized by buckling of all columns on the perime-
ter. The number of circumferential halfwaves correspond to the number columns
(silo columns buckled inward or outward) (Fig. 8). Buckling form in the third
range was appropriate for buckling of orthotropic shell. On one of circumferential
halfwave there were more than one silo columns so the number of circumferential
halfwaves was not depended from the number of silo columns. Calculated columns
spacing ds,max (between the range 2 and 3) was compatible to described by Eu-
rocode 3 [9]. However, the denition of the border between ranges 1 and 2 has not
been determined in Eurocode 3 approach. The number of halfwaves along the silo
height was equal three in all of the analyses cases of the silo column spacing.

4.2. Calculation of simplified silo model

The buckling load factor LBA for the simplied silo model with dierent distance
of columns was consistent with the result of the whole structure only in the rst
range of the column number (from 0 to 12 columns) or for a sparse column spacing
(Fig. 7). Buckling mode of the simplied model was constant and independent
of the column spacing (Fig. 9). At a certain spacing of columns the results of
the analyzed models dier signicantly and moreover, the results of the simplied
model gave overestimated buckling resistance of the silo (in range 2).
The buckling resistance of the simplied silo model as compared to the Eurocode
3 [9] approach (Fig. 7) were about 9 times higher. Application of the proposed in
[11] stiness of the column elastic foundation to the Eurocode 3 [9] formula (Eq.
(1)) was closer to FEM analysis than the formula currently recommended in code.
Additionally solution gave safe estimation of the silo buckling capacity.
118 Rejowski, K. and Iwicki, P.

A)

B)

Figure 8 Buckling modes of the whole silo model: A) silo 3D view and B) horizontal cross section
of the silo

A) B)

Figure 9 Buckling modes of simplied silo model: A) view and B) horizontal cross section

5. Conclusions
Some conclusions can be drawn from our stability FE studies for a dierent models
of the silos made from corrugated curved sheets strengthened by vertical columns:

The silo design according to the orthotropic shell theory in Eurocode [9] is
more realistic (Eqs (3) and (4)) than the design based on the method con-
sidering the column resting on the elastic foundation (Eqs (1) and (2)). The
Eqs. (1) and (2) are largely too conservative as compared to the FE results.
The proposed modication of the column elastic foundation stiness gave
results comparable to the FE numerical analysis (dierences about 20%) and
safe estimation of the silo buckling capacity.
 Buckling Analysis of Cold Formed Silo Column 119

The proposed simplied model consisting of a part of the silo wall was com-
parable with the whole silo FEM analysis only in the range 1 of the column
distance.

Eurocode 3 method A based on the orthotropic shell was about 30% lower
than the results obtained by whole silos FEM analyses.

The limit of the applicability of the simplied model should be determined

(between range 1 and 2).

Our studied will be continued in order to verify the proposed methods for the
calculation buckling strength of silo with dierent geometry.
In our future research a simplied silo model within range 2 and the inuence
of the horizontal pressure will also be investigated and the local buckling of the
columns will be taken into account.
Acknowledgements
The nancial support of the Polish National Research Centre NCN in the frame of
the Grant No 2011/01/B/ST8/07492 Safety and optimization of cylindrical metal
silos containing bulk solids with respect to global stability and of the Polish
Nation-al Centre for Research and Development NCBR in the frame of the Grant
POIG.01.03.01-00-099/12 Innovative method of dimensioning and construction of
large industrial silos made from corrugated sheets is gratefully acknowledged. The
numerical calculations were performed using the computing resources of CI TASK
at Gdask University of Technology.

References
[1] Teng, J. G. and Rotter, J. M.: Buckling of Thin Metal Shells, Taylor & Francis,
London, 2004.
[2] Singer, J., Baruch, M. and Harari, O.: On the stability of eccentrically stiened
cylindrical shells under axial compression. Int. J. Solids Struct., 3, 445470, 1967.
[3] Singer, J., Arbocz, J. and Babcock, C. A.: Buckling of imperfect stiened
cylindrical shells under axial compression, AIAA J., 9 (1), 6875, 1971.
[4] Nemeth, M. P.,: Buckling analysis for stiened anisotropic circular cylinders based
on Sanders nonlinear shell theory, NASA Technical Report, TM-2014-218176, 2014.
[5] Wojcik, M., Iwicki, P. and Tejchman, J.: 3D buckling analysis of a cylindrical
metal bin composed of corrugated sheets strengthened by vertical stieners, Thin
Walled Structures, 49, 947963, 2011.
[6] Iwicki, P., Wojcik, M. and Tejchman, J.: Failure of cylindrical steel silos com-
posed of corrugated sheets and columns and repair methods using a sensitivity anal-
ysis, Eng. Fail. Anal., 18, 20642083, 2011.
[7] Rotter, J. M. and Sadowski, A. J: Cylindrical shell bending theory for orthotropic
shells under general axisymmetric pressure distributions, Eng. Struct., 42, 258265,
2012.
[8] Iwicki, P. Rejowski, K. and Tejchman, J.: 2015. Stability of cylindrical steel silos
composed of corrugated sheets and columns based on FE analyses versus Eurocode 3
approach, Eng. Fail. Anal., 57, 444469, 2015.
[9] EN1993-4-1, Eurocode 3: Design of Steel Structures, Part 41: Silos, Tanks and
Pipelines Silos, CEN, Brussels,2009.
120 Rejowski, K. and Iwicki, P.

[10] Sondej, M., Iwicki, P., Tejchman, J. and Wojcik, M.: Critical assessment of
Eurocode approach to stability of metal cylindrical silos with corrugated walls and
vertical stieners, ThinWalled Structures, 95, 335346, 2015.
[11] Rejowski, K. and Iwicki, P.: Simplied stability analysis of steel cylindrical silos
with corrugated walls and vertical columns, XIII International Conference on Metal
Structures, Zielona Gora (In press), 2016.
[12] EN1991-1-4, Eurocode 1: Basis of Design and Action on Structures, Part 4: Actions
in Silos and Tanks, CEN, Brussels, 2008.
[13] Rejowski, K. and Iwicki, P.: Buckling analysis of cold formed silo column, Stability
of Structures XIV th Symposium, Zakopane, 9798, 2015.
[14] ABAQUS, Theory Manual, Version 6.8, Hibbit, Karlsson & Sorensen Inc., 2008.