D.Kober Lectures 3+4 Silos PDF

Silos
EN 1991-4
EN 1992-3
EN 1998-4
www.bing.com
D. Köber - Special Reinforced Concrete Structures – Lecture 1 1/69

Silos. Table of Content
1. Seismic input
1. Seismic design
2.1 Dynamic pressure

2.2 Behavior factor
2.3 Limit state verifications
2.4 Anchorages
3. Behavior of silos calculated acc. To the Janssen equations
4. Specific requirements
5. Practical calculation
6. Damage to silos

Silos. Seismic input
Source: Munich Re (2006)
- earthquakes are on the third place concerning the worldwide losses

natural hazards produce

- earthquakes are waves generated by a soil rupture or by a tectonic

movement, that shake the ground surface
- inertial forces are generated in structures and have to be transmitted
back to the foundation soil
global seismic weight of the

coefficient structure
structure dependent soil dependent

factors factors

soil dependent factors ag

soil dependent factors EN 1998-1 S, Tc

soil dependent factors P100-1 Tc

soil dependent factors P100-1 Tc

structure dependent factors
λ = 1 for silos
T [s]
γ m

soil and structure dependent factors q
F lower forces more deformation

(demand and remanent)
q higher forces less deformation
(demand and remanent)
Δ
q
 for silos the standard gives q<1,5 unless not proofed
 real design showed that the foundation soil can account for
plastic deformations for q 4,0

 how can a standard give us the seismic input for all:

- regular structures
- soil conditions
- earthquake regions
 with the help of response spectra

(measure / period, frequency)

response spectrum
m2 mn
m1
k1 k2 …. kn
T1 T2 Tn

 earthquake forces
- are an estimation
- must be supported and transmitted to the foundation
soil in a redundant manner
- force - deformation
“Enough resistance,
Δ F high ductility” (Bertero)
F Δ

 energy is absorbed foundation soil

 energy is consumed dissipative elements
plastic hinges

 silos are stiff structures no deformable regions

no plastic hinges elastic seismic response ?
 elastic seismic response high forces which cannot
be founded
 solution: plastic deformations in the ground lower
force level (behavior factor > 1)
F
3 2
1

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Seismic design
 seismic response spectrum design earthquake

ULS
 SLS design earthquake reduced by a factor ν
(equal to 0,5 for importance class I and II;
equal to 0,4 for importance class III and IV)
 for silos the consideration of the soil – structure
interaction is mandatory
 damping coefficient for ULS:
- 5% for the structure
- 10% for granular stored material
- 0,5% for liquid stored material (tanks)
- 25% for the foundation soil

 for seismic design we should consider the upper fractile

for the self weight of the stored material, γ
 because the stored material may move under seismic

action, a supplementary pressure appears on the
structural elements of the silo cell, than the pressure
given by the mass inertia of the stored material
 the self weight of the stored material should be affected

by a coefficient ψE,i = 0,8 for the seismic load combination

 double symmetric silo structures may be calculated

separately for each main direction (horizontal seismic
input in one direction + vertical seismic input)
 unsymmetric silos should be calculated for

simultaneously horizontal seismic input in both main
directions + vertical seismic input

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Seismic design. Dynamic pressure
 if we model realistic the characteristics of the stored

material and its dynamic response, than the seismic
action on the silo walls may be considered as a
supplimentary pressure:
- reference pressure value

θ – angle between
the radius and the
seismic direction
(0°….360°)

 the reference pressure value for vertical silo walls has to

be calculated at different heights x:
α(z) – response acceleration of the

silo at height z / g
for vertical walls
hb – overall silo heigt (with funnel)
dc – internal diameter of the silo cell

 the reference pressure value for the funnel has to be

calculated at different heights x:
α(z) – response acceleration of the

silo at height z / g
for vertical walls
hb – overall silo heigt (with funnel)
dc – internal diameter of the silo cell
β – funnel slope

 the sum of the static wall pressure and the seismic

supplement should be always positive
 otherwise the supplementary pressure from the seismic

action will be neglected

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Seismic design. Behavior factor
 silos with structural walls until the foundation should be

computed with low behavior factors (low ductility class,
DCL) unless a soil-structure interaction is considered.
 silos with flexible support structure may be computed with

higher behavior factors, depending on the structural type
and material.
 plan irregularities should be accounted for by reduction of

the behavior factor with a coefficient of 0,7

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Seismic design. Limit state verifications.
SLS
 silos should respect the regular criteria for reinforced
concrete structures
ULS
 sliding and overturning should be excluded
 at the border between silo and foundation the shear
sliding forces should be determined under consideration
of the vertical seismic component
 slight sliding is accepted as long as it is compatible with
the connecting structural elements and the pipes
 lifting of the foundation is accepted if it is considered in
the design

Silos. Seismic design. Limit state verifications.
ULS
 internal forces for the walls should be less than the wall
capacity under nonseismic actions:
limit state for bending + axial force

limit state for shear force
 design will follow normal RC regulations

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Seismic design. Anchorages
 anchorages should be designed in order to assure elastic

behavior
 they should have enough ductility in order to avoid fragile

failure
 for the anchorage design a safety factor of minimum 1,25

should be considered
 if the anchorage is part of a dissipative mechanism, than

sufficient ductility should be provided

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Behavior of silos calculated acc. to the Janssen equations
 1895, when Janssen came up with his equations, silos

with H = 20 m, D = 3÷5 m and vunloading = 10÷20 tons/hour
were built.
 low unloading velocities → low horizontal pressures →
horizontal reinforcement is almost constructive
 starting in 1910 the grain production was rising especially
in Canada and the USA
 at that time shortly silos with H = 40÷50 m, D = 7÷9 m
and vunloading = 150 tons/hour were built
 at those silos soon damage was observed

 cracks on 2/3 H from the bottom, having widths of some

mm were observed and even cell crush
 retrofit of silos is difficult to achieve and cost intensive
 experimental studies were

conducted especially in the former
URSS, France, Germany and the
USA
 all experiments showed that the
Janssen theory is valid for low silo 2/3 H
heights, low diameters and low
unloading velocities

 why?
 because at low unloading velocities the stored material is

almost static and the minimum required reinforcement
covers the needs
 conclusion at that time: the Janssen theory is valid for

every H and D assuming static storing!
 at unloading, depending on the unloading velocity, the
static equilibrium is disturbed and the stored material
behaves as a semifluid, having a nonlinear flow because
of the unhomogenous friction between the stored material
particles

 during 3 decades experimental studies couldn’t come up

with a solution, because:
- they studied models at reduced scale which could
not reproduce phenomena that appear in a real cell
- at real scale pressure indicators were installed on a
vertical line, changing the wall stiffness in that
region and influencing the real behavior
3m
var. = (0,5÷1,0)
po,Janssen
0.6 m

 German experiments showed also higher pressures in

the upper silo part
→1,6 po,Janssen
envelope
 no one could explain the great variation of experimental

results

 the decision was to rise the Janssen pressure on the first

2/3 of the silo height:
p0calcul = a·p0,J
a = 1,5 ÷ 2,5 variation coefficient
1/3 H
2 po,Janssen
2/3 H
po,Janssen

 when the real-time measurement technology was

discovered, the Russians made a full scale testing with
pressure measurements on 8 vertical axes (simultaneous
measurement in 80 points)
 just than the pressure distribution in the silo cell for great
heights, diameters and unloading velocities could be
determined
 in every point a variable pressure was registered

 maximum and minimum values occurred at different time
steps in different points
p0
p0,med
unloading
p0,J
t
 at unloading the stress distribution in the cells is highly
different from the stress distribution during the static
storing
 at unloading stress variations occur along the perimeter
as well as along the cell height (mean value 2p0,J)

 if along the height a safe side rising of the Janssen

pressure may be assumed, the variation along the
perimeter of the cell is random
 a rising of p0,J to 2p0,J as mean value doesn’t cover the
peaks that occur during the unloading process
distribution pressure distribution

depending on after neglecting p0,med
unloading
H/D
p0,med
unloading
p0,J
p0,measured

 p0,med produces only axial forces in the cell walls:

unloading
N = p0,med · cell radius

unloading
 the pressure variations produce in a circular cell an

ovalisation due to low axial forces but high bending
moments
 each point on the silo cell experiences

± bending moments v
 real internal forces should be considered
for design

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Specific requirements
 central flow channels may cause segregation in grain

silos. The grains are heavier and tend to unload first. On
the lateral surface of the silo cell the smaller parts remain.
Two qualities of flour result from the same silo.
 that is why grain silos are provided with 4

funnels

 for a proper exploatation process, air is introduced into

the silo cells in order to avoid deterioration of the stored
material
 therefor pipes are introduced into the silo cell over the
funnel. The pipes go through the silo cell up to the stored
material surface
 air introduced to the pipes by a compressor is released in
the silo cell through openings in the pipes
 the horizontal air pressure varies linear from zero to
maximum
 air introduction is symmetric along the perimeter and
produces only axial forces in the silo cell

p0,air
 pipes have regular diameter of 60 ÷ 80 cm

 silos in the cement industry need homogeniosation of the

stored material (powder mixture)
 the stored material becomes a semi-fluid
 the stored material pressure becomes liniar, as for fluids
 because we have some friction, the maximum value is
0,6γH
 it is a big value, so mixture
silos can’t be too high
semifluid
0,6γH

 temperature variation may cause supplementary internal

forces:
day-night – ΔT≈15°
summer-winter - ΔT≈ 40÷50°
 at rising temperature, dilatation processes occur

 the stored material level drops
 variation without supplementary pressure
in the silo cell
R
R+ΔR

 when the temperature drops, the silo cell is shrinking

 phenomenon especially dangerous for steel silos with
bolt connections
 the stored material gives a passive opposition to the
shrinkage tendency
 a triaxial loading occurs in the stored material because it
cannot rise vertically as a liquid, due to friction
 the greater the silo height, the less can the
stored material at the bottom of the cell
move
Δpt
R
R-ΔR

 variable triaxial loading in radial direction
p0+Δpt
p0
R
 part of the stored material gets compressed in the

neighborhood of the silo wall
 until the compression limit is reached
 than high pressures are registered
Δpt
R
R-ΔR

 under seismic loading a so called “wave” effect occurs

 it is especially dangerous for tanks:
 in silos, due to friction between the stored material

particles, the wave effect is reduced
 practice showed that the pressure variation is greater
during the unloading process than under seismic action

 in order to keep the stored material under control,

thermometers are suspended inside the silo cell (in the
center and on two diameters)
 the thermometers are steel wires which are suspended
from the slab over the cells
 the wires have temperature sensors
 a friction between the wires and the stored material
occurs
 bending in the slab over the cells appears
due to wires tension

 stresses from the technological process should be

accounted for when designing a silo
 conveyor belts and other machines should be considered
 the elevator tower vibrates continuous and in time
shrinking and creep of concrete occur
 a battery of silo cells may encounter uneven compaction
on the foundation soil, especially due to unsymmetrical
loading of the cells
 at least until the foundation soil gets used to the weight of
the silo, the loading process must be uniform and
controlled: ¼, ½, 2/3, full. This process lasts 1 year and
avoids remanent soil deformations from the fundamental
load combination
 after the foundation soil has reached its maximum
compaction level, no controlled loading is needed any
more
1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Practical calculation
 Janssen:
horizontal pressure phf
vertical pressure pvf
friction pressure pwf
 only the horizontal pressure needs correction due to
unloading

 at unloading supplementary:
horizontal symmetric pressure phf1 = a1·phf
(acting along a ring)
horizontal unsymmetric pressure phf2 = a2·phf

(to account for ovalisation)
s=0,2dc 0,25dc phf1
phf2 s=0,2dc

 phf1 may act everywhere along the silo height
 phf2 may act everywhere along the silo height and along
the cell perimeter
 experimental studies have shown that the pressure

variation at unloading is totally random
 so we may not hope for an analytical solution in order to

determine the real pressures at unloading
 the assumption based on phf1 and phf2 is considered to

give safe side results

 the coefficients a1 and a2 depend on the ratio hb/dc

hb/dc
coeff. 10 5 2,5 1,67 1,25 0,83 0,625
a1 1,8 1,5 1,2 0,9 0,7 0,5 0,3
a2 1,25 1,0 0,50 0,25 0,12 0,06 0,03
 for hb/dc >10 we have an infinite high

cell
 hb/dc=0,625 is at the border to bunker
(a silo with neglect able friction and
just active pressure)

 for unloading, the horizontal pressures can be

determined as follows:
- phf,total = phf + phf2

RC cells, steel cells with stiffenings (stiff in the
horizontal plane)
- phf,total = phf + phf1

cells without stiffness in the horizontal plane, which
take over the shape of the stored material at each
moment and are subjected only to axial tension

1. Seismic input
1. Seismic design

2.2 Behavior factor
2.4 Anchorages
6. Damage to silos

Silos. Damage to silos








D.Kober Lectures 3+4 Silos PDF

Uploaded by

Copyright:

Available Formats

D.Kober Lectures 3+4 Silos PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

D.Kober Lectures 3+4 Silos PDF

Uploaded by

Copyright:

Available Formats

Silos

D. Köber - Special Reinforced Concrete Structures – Lecture 1 1/69

2.1 Dynamic pressure

3. Behavior of silos calculated acc. To the Janssen equations

D. Köber - Special Reinforced Concrete Structures – Lecture 1 2/69

Source: Munich Re (2006)

- earthquakes are on the third place concerning the worldwide losses

D. Köber - Special Reinforced Concrete Structures – Lecture 1 3/69

- earthquakes are waves generated by a soil rupture or by a tectonic

global seismic weight of the

structure dependent soil dependent

D. Köber - Special Reinforced Concrete Structures – Lecture 1 5/69

soil dependent factors ag

D. Köber - Special Reinforced Concrete Structures – Lecture 1 6/69

soil dependent factors EN 1998-1 S, Tc

D. Köber - Special Reinforced Concrete Structures – Lecture 1 7/69

soil dependent factors P100-1 Tc

D. Köber - Special Reinforced Concrete Structures – Lecture 1 8/69

soil dependent factors P100-1 Tc

D. Köber - Special Reinforced Concrete Structures – Lecture 1 9/69

structure dependent factors

D. Köber - Special Reinforced Concrete Structures – Lecture 1 10/69

soil and structure dependent factors q

F lower forces more deformation

D. Köber - Special Reinforced Concrete Structures – Lecture 1 11/69

 how can a standard give us the seismic input for all:

 with the help of response spectra

D. Köber - Special Reinforced Concrete Structures – Lecture 1 12/69

D. Köber - Special Reinforced Concrete Structures – Lecture 1 13/69

D. Köber - Special Reinforced Concrete Structures – Lecture 1 14/69

 energy is absorbed foundation soil

D. Köber - Special Reinforced Concrete Structures – Lecture 1 15/69

 silos are stiff structures no deformable regions

D. Köber - Special Reinforced Concrete Structures – Lecture 1 16/69

2.1 Dynamic pressure

3. Behavior of silos calculated acc. To the Janssen equations

D. Köber - Special Reinforced Concrete Structures – Lecture 1 17/69

 seismic response spectrum design earthquake

D. Köber - Special Reinforced Concrete Structures – Lecture 1 18/69

 for seismic design we should consider the upper fractile

 because the stored material may move under seismic

 the self weight of the stored material should be affected

D. Köber - Special Reinforced Concrete Structures – Lecture 1 19/69

 double symmetric silo structures may be calculated

 unsymmetric silos should be calculated for

D. Köber - Special Reinforced Concrete Structures – Lecture 1 20/69

2.1 Dynamic pressure

3. Behavior of silos calculated acc. To the Janssen equations

D. Köber - Special Reinforced Concrete Structures – Lecture 1 21/69

 if we model realistic the characteristics of the stored

- reference pressure value

D. Köber - Special Reinforced Concrete Structures – Lecture 1 22/69

 the reference pressure value for vertical silo walls has to

α(z) – response acceleration of the

D. Köber - Special Reinforced Concrete Structures – Lecture 1 23/69

 the reference pressure value for the funnel has to be

α(z) – response acceleration of the

D. Köber - Special Reinforced Concrete Structures – Lecture 1 24/69

 the sum of the static wall pressure and the seismic