Steel Strengthening - Elyas

Table of content of printouts:
Introduction
Materials and Properties of Polymer Matrix Composites
Mechanics of a Lamina
Laminate Theory
Ply by Ply Failure Analysis
Externally Bonded FRP Reinforcement for RC Structures: Introduction and Basics
Flexural Strengthening
Strengthening in Shear
Column Confinement
FRP Strengthening of Masonry
FRP Strengthening of Metallic Structures
FRP Strengthening of Timber Structures
Design of Flexural Post-Strengthening of RC: Swiss Code 166 and Other Codes/Guidelines
Design of FRP Profiles and all FRP Structures
An Introduction to FRP Reinforced Concrete
Monitoring and Testing of Civil Engineering Structures
Composite Manufacturing
Testing Methods
Steel Strengthening
Fibre Composites, FS15
Elyas Ghafoori
Strengthening of Metallic Members using CFRP Laminates
Elyas Ghafoori
Dr. sc. ETH Zrich, M.Sc. Mech.-Eng. EPFL
Research Scientist
Empa, Swiss Federal Laboratories for Materials Science and Technology
berlandstrasse 129, 8600 Dbendorf, Switzerland
Tel +41 58 765 4935
Researchgate: https://www.researchgate.net/profile/Elyas_Ghafoori
Perosnal Webpage: http://www.empa.ch/plugin/template/empa/357/*/---/l=2/uacc=ghe
E-mail: elyas.ghafoori@empa.ch
Steel Strengthening
Elyas Ghafoori
Outlines
Introduction & Motivation

Lateral Torsional Buckling (LTB) Strengthening
Fatigue Strengthening of
Healthy Metallic Members
Cracked Metallic Members
Case Study: Fatigue Strengthening of Mnchenstein Railway Bridge
New Projects
Master Thesis Proposals
Steel Strengthening
Elyas Ghafoori
Introduction
Market:
Europe:
22% bridges are metallic
70% are older than 50 years
Switzerland:
Swiss Federal Railways (SBB) has 6050 railway bridges
25% of bridges older than 80 years are metallic riveted
Problems in Metallic Bridges:

Major problem: insufficient fatigue crack safety
Need an upgrade to carry larger loads and more traffic
Most commonly used structural metals:

Steel, wrought irons, cast irons, Aluminum and
Chajes et al. Fracture: Field testing of the I-95 bridge. In Third Annual Bridge Workshop: Fatigue and
Fracture; Center of Innovative Bridge Engineering: Ames, IA,USA, 2004
Steel Strengthening
Elyas Ghafoori
Introduction
Kuehn et al. Assessment of Existing Steel Structures: Recommendations for Estimation of

Remaining Fatigue Life; the Publications Office of the European Union: Luxembourg, 2008
Kuehn et al. Assessment of Existing Steel Structures: Recommendations for Estimation of

Remaining Fatigue Life; the Publications Office of the European Union: Luxembourg, 2008
Fisher, J.W. Fatigue and Fracture in Steel Bridges; Wiley-Interscience: Hoboken, USA, 1984
Daniel Hoan Memorial Bridge, Milwaukee, Wisconsin, Failure on the 13th of December 2000
Steel Strengthening
Elyas Ghafoori
Introduction: Why CFRP Laminates?

Traditional Strengthening Solutions:
Steel: heavy
CFRP:
Excellent fatigue behavior
High fatigue-to-weight ratio
Classifications of the CFRP laminates according to their

Youngs modulus relative to that of steel:
Steel Strengthening
Laminate type
Laminate modulus
Low modulus (LM)
< 100 GPa (ECFRP < 0.5 Esteel)
Normal modulus (NM)
100 200 GPa (0.5Esteel ECFRP < Esteel)
High modulus (HM)
200400 GPa (Esteel ECFRP < 2Esteel)
Ultra-high modulus (UHM)
400 GPa (ECFRP 2 Esteel)
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M. Normal, high and ultra-high modulus CFRP laminates for bonded and un-bonded strengthening of steel beams. Materials and Design, 2015. 67: p.
232243.
Introduction: Change in CFRP Youngs Modulus
S&P 150/2000
S&P 200/2000
Carbolam THM 450
E=440 GPa
HM
= 178.8+1.482
R = 0.9999
2500
350
250
E=205 GPa
200
150
HM
NM
E=165 GPa
NM
2000
= 399.9+11.82
R = 0.9998
1500
50
0
2
8
10 12
Strain [mm/m]
14
16
18
The measured Youngs modulus of the NM, HM, and UHM

CFRP laminates as a function of the applied strain. The
square markers show the Youngs moduli provided by the
manufacturers.
Steel Strengthening
C
B
= 143.3+1.22
R = 0.9999
D
1000
E
D B
500 C
B
100
UHM
UHM
300
Stress, [MPa]
Young's modulus, ECFRP [GPa]
450
400
3000
NM
UHM
HM
500
8
10
12
Strain, [mm/m]
14
16
18
20
S&P 150/2000
A (E = 141.2 GPa)
B (E = 152.4 GPa)
C (E = 156.8 GPa)
D (E = 162.3 GPa)
E (E = 171.1 GPa)
S&P200/2000
A (E = 168.6 GPa)
B (E = 187.9 GPa)
C (E = 194.5 GPa)
D (E = 202.7 GPa)
E (E = 208.8 GPa)
Carbolam THM 450
A (E = 385.5 GPa)
B (E = 379.3 GPa)
C (E = 422.3 GPa)
D (E = 438.5 GPa)
E (E = 435.0 GPa)
Fu_UHM
Fu_HM
Fu_NM
The measured Youngs modulus for the NM, HM, and UHM CFRP
laminates at different strain levels, indicating a non-linear elastic behavior
for the CFRP laminates.
Elyas Ghafoori
232243.
Introduction: Some of Typical Failure Modes of Steel Beams

Bonded with CFRP Plate
Steel Strengthening
Elyas Ghafoori
Introduction: Possible Failure Modes of CFRP-toConcrete/Steel Bonded Joints
Concrete
Concrete failure
The main difference between FRPsteel and FRPconcrete bonded joints is that in the former, failure will likely occur in the
adhesive layer and in the latter failure is expected to occur in the concrete. Therefore, by providing an adequate bond length,
the optimal strength of a bond joint is dependent on the fracture energy of the adhesive for the former and the fracture
energy of the concrete for the latter.
In the FRP-strengthened steel structures, interfacial failure should happen within the adhesive layer in the form of cohesion
failure to maximize the effectiveness of FRP strengthening.
Inappropriate surface preparation of the steel substrate prior to the bond application will result in an adhesion failure at the
steel-to-adhesive interface.
Steel Strengthening
Elyas Ghafoori
Steel Strengthening
Elyas Ghafoori
Steel Beam Strengthened by a Prestressed Bonded Plate
q
tf
Beam
x,u
y,w
tw
bp
bf
Lp
Ms(x)
Beam
Ns(x)
Ns(x)+dNs(x)
dx
ta tp
Vs(x)
(x)
Beam
(x)
Vs(x)+dVs(x)
(x)
Adhesive layer
(x)
(x)
Prestressed plate
(x)
Np(x)
Steel Strengthening
Ms(x)+dMs(x)
Plate
Np(x)+dNp(x)
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M., Zhao X.L., Nussbaumer A., Fontana M. Fatigue design criteria for strengthening metallic beams with bonded CFRP plates. Engineering Structures,
2015. 101: p. 542-557
tf
Beam
x,u
bp
bf
y,w
tw
Lp
Ms(x)
Beam
Ns(x)
Ns(x)+dNs(x)
dx
ta tp
Vs(x)
(x)
Beam
(x)
Vs(x)+dVs(x)
(x)
Adhesive layer
(x)
(x)
Prestressed plate
(x)
Np(x)
Ns x N p x N x
s x
p x
dus x
du x hM s x
sN x sM x
,
dx
dx
2 Es I s
N
s
du p x
dx
N p x
E p Ap
N p x N p x N0 ,
Ga
u p x us x ,
ta
d x Ga du p x dus x
,
dx
ta dx
dx
Steel Strengthening
Ms(x)+dMs(x)
Plate
Np(x)+dNp(x)
d x Ga N p x N 0 hM s x N s x
,
dx
ta
E p Ap
2 Es I s
Es As
(1)
(2) & (3) into (6) =>
(2)
d 2 x Ga 1 dN p x
h dM s x
1 dN s x
.
dx 2
ta E p Ap dx
2 Es I s dx
Es As dx
(3)
dN s x dN p x
bp x ,
dx
dx
Moment equilibrium: dM s x Vs x bp h x
dx
2
Force equilibrium in x direction:

(4)
(5)
(6)
d 2 x Ga bp 1
1
h2
h
VT x .
x K
2
dx
ta Es I s E p I p 4 I s Es
2 Es I s
d 2 x
x x .
dx 2
(7)
(8)
(9)
(10)
(11)
(12)
Elyas Ghafoori
2015. 101: p. 542-557
P or [Pmin - Pmax]
b
b
I-Beam
x,u
tf
Lp
a
y,w
h ta
tw
bp
bf
tp
Beam strengthened by the bonded CFRP laminate in a four-point bending set-up.

By applying the boundary conditions for the above four-point bending set-up:
G N
hPa x

0
a
e m1P 1 e cosh x
ta E p Ap 2 Es I s
x
x
hPa
Ga N 0
m
P
sinh
1
t E A
2
E
I
a
p
p
s
s
Ga bp 1
1
h2
;
ta Es I s E p I p 4 Es I s
From (1) & (7) =>
N x
m1
0 x ba
(13)
b a x Lp / 2
Ga
h
2
2ta Es I s
bp d x
GN
m1 2 M T x a 0
2
dx
ta E p Ap
(14)
(15)
(16)
Stress in bottom flange:
x MT x
GN
h bp h 2
1 d x
2

m1 2 M T x a 0
2 I s 4 I s As dx
ta E p Ap
Steel Strengthening
(17)
Elyas Ghafoori
2015. 101: p. 542-557
P or [Pmin - Pmax]
b
tf
I-Beam
x,u
Lp
bp
bf
y,w
h ta
tw
tp
A simply supported beam with a free span of L=1200 mm and a

CFRP plate length of Lp=920 mm (i.e., a=140 mm) in a four-point
bending set-up:
b f 65 mm, t f 6.2 mm, tw 4.4 mm, h 120 mm, t p 1.4 mm, bp 50 mm,
ta 1 mm, Es 210 GPa, Ga 1040 MPa.
1800
1500
1200
900
600
300
HM CFRP
Shear & Bending

1500
1200
900
P=40 kN
Experiment: NM CFRP
Model: NM CFRP
Experiment: HM CFRP
Model: HM CFRP
Experiment: UHM CFRP
Model: UHM CFRP
600
300
0
0.1
0.2
0.3
0.4
0.5
0.1
0.2
0.3
0.4
0.5
x/Lp
x/Lp
The calculated and measured strains along the CFRP laminates for
the beam strengthened by the HM CFRP at actuator load levels of
P=15 kN, 30 kN and 51 kN (within the elastic domain).
Steel Strengthening
Bending
Anchorage
Modeling: 15 kN
Experiment: 15 kN
Modeling: 30 kN
Experiment: 30 kN
Modeling: 51 kN
Experiment: 51 kN
Strain in CFRP (micro-strain)
1800
The calculated and measured strain along the CFRP laminates for the
retrofitted beams with the NM, the HM and the UHM CFRP at an
actuator load level of P=40 kN (within the elastic domain).
Elyas Ghafoori
2015. 101: p. 542-557
Actuator
LVDT
Bonded CFRP laminate

Supports
40
Experiment: 0% pre-stress in NM CFRP

Model: 0% pre-stress in NM CFRP
Experiment: 0% pre-stress in HM CFRP
Model: 0% pre-stress in HM CFRP
Experiment: 0% pre-stress in UHM CFRP
Model: 0% pre-stress in UHM CFRP
35
6000
30
P=30 kN
4000
NM CFRP
Model: 0% pre-stress
Experiment: 0% pre-stress
2000
0
0
0.1
0.2
0.3
0.4
Distance from the laminate end (m)
Force in actuator (kN)
8000
25
20
15
10
5
0
-700
The calculated and measured strains along the CFRP

laminates for the beams strengthened by the NM CFRP with
0%, 20% and 40% pre-stress levels subjected to an actuator
load level of P=30 kN.
Steel Strengthening
-350
0
350
700
1050
Strain in steel bottom flange (micro-strain)
1400
The strain in the bottom flange of the specimens strengthened by the

NM, the HM and the UHM CFRP laminates with different pre-stress
levels while the actuator load, P, increases from 0 to 40 kN.
Elyas Ghafoori
2015. 101: p. 542-557
LVDT
Actuator
Bonded CFRP laminate

Supports
100
80
90
70
80
60
Load (kN)
Load (kN)
70
60
Experiments of steel beams with
y=355 MPa
50
40
50
Experiments of steel beams
with y=235 MPa
40
30
30
20
20
B1: control (unstrengthened) beam

B2:bonded with the plate @ 27% prestress
Presented analytical method for B2
10
0
0
10
15
20
25
Mid-span deflection (mm)
30
B4: bonded with the plate @ 40% prestress
10
Presented analytical method for B4

0
35
Comparison of load-deflection behaviors of steel beams with y=355

MPa, one unstrengthened and one strengthened with a 27% prestressed
bonded CFRP plate, both loaded in a four-point bending set-up.
Steel Strengthening
B3: control (unstrengthened) beam
10
15
20
25
Mid-span deflection (mm)
30
35
Comparison of load-deflection behaviors of two steel beams

with y=235 MPa, one unstrengthened and one strengthened
with a 40% prestressed bonded CFRP plate, both loaded in a
four-point bending set-up
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M., Flexural and interfacial behavior of metallic beams strengthened by prestressed bonded plates, Composite Structures, 101 (2013), 22-34.
Strengthening against Lateral Torsional Buckling
Steel Strengthening
Elyas Ghafoori
Definition of Lateral Torsional Buckling (LTB)

The LTB failure is often triggered in slender beams, which do not have sufficient lateral supports, due to eccentricities,
and can occur at load levels that are below yield capacity.
These eccentricities, in reality, can be due to the geometrical imperfections of the beam itself or the position of the loads.
The eccentricity generates a bending moment about the longitudinal axis, which displaces the compression flange
laterally away from the loading plane, while the tension flange tends to keep the beam straight, and thus, the beam cross
section is twisted.
This twisting in combination with the lateral displacement of the beam is called the LTB failure and could occur well
before the yielding capacity of the steel cross section is reached.
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M. Lateral-torsional buckling of steel I-beams retrofitted by bonded and un-bonded CFRP laminates with different pre-stress levels: experimental and
numerical study. Construction and Building Materials, 2015. 76: p. 194206.
Mechanisms of Strengthening against LTB

Two retrofit mechanisms:
Increasing the out-of-plane stiffness of the beam using UHM CFRP laminates:
For the LTB failure, the specimen buckles out of the plane of the flexural loading, and the CFRP laminates can affect the buckling
capacity of the retrofitted beams by stiffening the steel cross section around the weak axis. Application of the ultra-high modulus
CFRP laminates increases the out-of-plane stiffness of the specimens, and consequently, the buckling strength of the beams
increases.
Applying tension to the top flange of the beam using pre-stressed CFRP laminates:
Whether the prestressed CFRP laminate leads to tensile or compressive stresses in the top flange depends only on the profile
geometry. Assuming that the prestressing is applied on the bottom surface of the bottom flange, the stresses in the top flange
caused by the axial force and by the bending moment can be calculated as:
A
p
top
ph / 2 h P
I
0 h 2.
I
2 A
A
Steel Strengthening
Condition for top flange to be in tension
Elyas Ghafoori
232243.
Important Notes on CFRP Strengthening against LTB

When
h 2.
I
A
, use of CFRP laminates with high pre-stress levels is NOT recommended! Instead, we can use UHM laminates.
F
P =X or 2X
+
Negative moment due to
CFRP stress level of X
Compression due to
CFRP stress level of X
+
Negative moment due to
CFRP stress level of 2X
P =X or 2X
More of cross-section is
under compression
Resulting
stresses
Moment due to
external loading
Compression due to
CFRP stress level of 2X
=
Moment due to
external loading
Higher probability
of buckling
Resulting
stresses
Stress distribution in the beam cross-section for the two different pre-stress levels of X % and 2X %.
Note: Application of CFRP laminates to the tension face of the steel beams increases the in-plane bending strength and also the lateral
buckling strength; however, the former increases more significantly. This arrangement could change the failure mode of the steel
beam from in-plane bending to the buckling failure mode after CFRP strengthening. This need to be check in advance!
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M. Lateral-torsional buckling of steel I-beams retrofitted by bonded and un-bonded CFRP laminates with different pre-stress levels: experimental and
numerical study. Construction and Building Materials, 2015. 76: p. 194206.
Fatigue Strengthening
Steel Strengthening
Elyas Ghafoori
Fatigue Strengthening
Healthy metallic members:
Stringer
Cross beam
Stringer
Cross beam
Rivet
Cracked metallic members:
Steel Strengthening
Elyas Ghafoori
Fatigue Strengthening of Healthy Metallic Members
Steel Strengthening
Elyas Ghafoori
Design criterion for fatigue strengthening of steel girders using bonded CFRP laminates
Fatigue Theory
Sy
Se
Sut
Material
properties
Yield strength
Fatigue endurance limit
Ultimate tensile strength
a, h
Stress amplitude a
Fatigue
life
Midrange
stress
m,h
a, h
Sy
1: No fatigue crack
max,2:h Fatigue crack may occur

3: Fatigue crack occurs
Stress
amplitude
min, h
Se
3
2
Stress
a
a
m
0
-
Safe
stresses at
Zone
root
of notch
-Sy
Time
max min
2
Steel Strengthening
Compression
Tension
Midrange stress m
max min
2
Elyas Ghafoori
Sy
Sut
Fatigue Theory
Before strengthening (A):
Stress
+
F
Steel beam
Sy
After strengthening (B):
by prestressed CFRP
Stress
Se
A
Safe
Zone
After strengthening (C):
Tension
Sy
Midrange stress m
Steel Strengthening
Sut
Stress
Compression
m Time
by increasing stiffness
-Sy
m Time
UHM-CFRP
Stress amplitude a
0
-
a
a
m Time
Elyas Ghafoori
Fatigue Theory
P
d x Ga bp
dx 2
ta
2
1
1
h
Es I s E p I p 4 Es I s
G N
hPa
0
x a
ta E p Ap 2 Es I s
flange
G
h
VT x
x a
t
2
E
I
a
s
s
Lp / 2
G N
m1 Pa 2 a 0

ta E p Ap
tf
h ta
tw
I-Beam
Lp
a
CFRP laminate
L
m1 P sinh e
h2
hPa
1
bp
2I s
4 I s As
bp
bf
Adhesive
Ms(x
)
interfacial shear stress along the CFRP plate

stress in beam bottom flange
N
force in CFRP plate
N0 the pre-stress level
Ga adhesive shear modulus
Note: Subcripts s and p refers to the steel and the CFRP plate
flange
Beam
Ms(x)+dMs(x)
Ns(x)
Ns(x)+dNs(x)
dx
Vs(x)
Vs(x)+dVs(x)
(x)
(x)
Adhesive layer
Ga bp 1
1
h
ta Es As E p Ap 4 Es I s
Ga
h
m1
2
2ta Es I s
2
Steel Strengthening
(x)
Np(x)
CFRP Plate
Vp(x)
(x)
Vp(x)+dVp(x)
Elyas Ghafoori
Np(x)+dNp(x)
tp
Fatigue Theory
Stress amplitude a
Sy
Se
A
B
Safe
Zone
a
Se
-Sy
Compression
m
S ut
Tension
Sy
Sut
Midrange stress m
flange
h2
G N
hPa
1 Lp / 2
bp
m1 Pa 2 a 0

2I s
ta E p Ap
4 I s As
b h2
haPa
1
Ga N 0
p
m1aPa 2
2 I s S e S e 4 I s As
ta E p Ap
Steel Strengthening
haPm
bp h 2
b d
1
Ga N 0
L /2
e p p
m1aPm 2
nb k
ta E p Ap
p f
2 I s Sut Sut 4 I s As
Elyas Ghafoori
Laboratory Verifications
65
32
Fatigue crack
grows in web
Crack initiates
from hole
50
Sudden failure
Fatigue crack growth

Debonded CFRP laminate
Steel Strengthening
Elyas Ghafoori
Design Example
4.0 m
1.2 m
1.6 m
Fd
1.2 m
Fd
Steel beam
Geometry of cross-section:
Steel grade: S235
h = 350 mm
w = 250 mm
fy = 235 MPa
fu = 360 MPa
ES = 210 GPa
tf = 16 mm
tw = 8 mm
dHole = 20 mm
Fatigue loading:
Steel Strengthening
Case A:
Case B:
Fd = 0.0 kN 84.0 kN
Fd = 0.0 kN 96.0 kN
Elyas Ghafoori
Design Example
Case A:
Fd = 0.0 84.0 kN
250
A: unstrengthened
B: Prestressed NM
C: Non-prestressed
150
100
A
B
100
Midrange stress m,h [MPa]
200
300
400
Youngs modulus
[GPa]
165.0
412.1
165.0
Plate thickness
[mm]
1.4
1.4
3.5
[%]
8.9
0.0
0.0
Prestressing level
Pre-stressing
level
Steel Strengthening
NM-CFRP
-100
UHM-CFRP
Prestressed
NM-CFRP
50
Unstrength
Unstrengthened
ened
Alternating stress a,h [MPa]
200
Elyas Ghafoori
Design Example
Case B:
Fd = 0.0 96.0 kN
250
A: unstrengthened
B: Prestressed NM
C: Non-prestressed
150
A
100
B
C
100
Midrange stress m,h [MPa]
200
300
400
Youngs modulus
[GPa]
165.0
950
165.0
Plate thickness
[mm]
1.4
1.4
7.8
[%]
20.3
0.0
0.0
Prestressing level
Pre-stressing
level
Steel Strengthening
NM-CFRP
-100
UHM-CFRP
Prestressed
NM-CFRP
50
Unstrength
Unstrengthened
ened
Alternating stress a,h [MPa]
200
Elyas Ghafoori
Conclusiones
The advantages of the proposed design approach:
1. it is a proactive strengthening approach,
2. it takes into account the combined effects of mean stress and alternating stress levels.
Two main fatigue retrofit mechanisms for healthy metallic members:
1. to decrease the mean stress level by using pre-stressed laminate
2. to decrease mean and alternating stresses proportionally by using ultra-high modulus laminate
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E., Motavalli M., Zhao X.L., Nussbaumer A., Fontana M. Fatigue design criteria for strengthening metallic beams with bonded
CFRP plates. Engineering Structures, 2015. 101: p. 542-557.
Fatigue Strengthening of Cracked Metallic Members
Steel Strengthening
Elyas Ghafoori
Crack Metallic Members

110
100
Specimen B2 (non-prestressed)
90
Specimen B5b (prestressed)
Crack length (mm)
80
70
60
50
40
30
20
10
0
0
5000
10000 15000 20000 25000 30000
A schematic view of a beam strengthened with the PBR system.

Number of cycles
5 times increase in fatigue life by using prestressed CFRP plate.
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E, et al., Fatigue strengthening of damaged steel beams using unbonded
and bonded prestressed CFRP plates, International Journal of Fatigue, 2012, 44 , pp. 303-315.
A comparison of the load-deflection curves for beam 3.
The strain distribution

along the cracked section of aFibre
beam
strengthened
Steel Strengthening
Composites, FS15
with a prestressed CFRP plate under different levels of bending moment.
Elyas Ghafoori
The deflection behavior at different crack lengths.
Fatigue Crack Arrest using Pre-stressed CFRP Plates
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E., Schumacher A., Motavalli M. Fatigue behavior of notched steel beams reinforced with bonded CFRP plates: Determination of prestressing level for crack arrest.
Engineering Structures, 2012. 45: p. 270-283.
Fatigue Crack Arrest using Pre-stressed CFRP Plates
Steel Strengthening
Elyas Ghafoori
Condition of Fatigue Crack Arrest

A crack will not grow when it is under compression.
Pre-stressed CFRP laminates can be used to apply compression to an existing crack in metal.
Based on fracture mechanics: K total K F K P 0
where K is the stress intensity factor
Steel Strengthening
Elyas Ghafoori
Can Pre-stressing Lead to Stiffness Increase ?
For cracked metallic members:

YES
For healthy metallic members:

NO*
* Note that strengthening using CFRP plates increases the stiffness of metallic beams, but prestressing does not always lead
to additional stiffness increase.
Steel Strengthening
Elyas Ghafoori
Case Study:
Fatigue Strengthening of Mnchenstein Railway Bridge
Steel Strengthening
Elyas Ghafoori
Bridge History
The Mnchenstein rail disaster on 1891 is historically the worst railway accident ever in Switzerland. The bridge had
been built in 1875 by Gustave Eiffel, who built the Eiffel Tower later in 1889.
Prof. Ludwig von Tetmajer, the first director of Empa, was commissioned to investigate the cause of this collapse. His
investigation led to modification of Eulers formula for buckling of slender bars.
2013
1891
Steel Strengthening
Elyas Ghafoori
Bridge History
Based on the verifications done by an engineering office*, the cross-beams of

Mnchenstein Bridge are the fatigue critical elements if further bridge serviceability after
2030 is intended.
Therefore, the goal of this pilot project is to demonstrate the capability and the
effectiveness of a pre-stressed un-bonded strengthening system to reinforce this bridge.
Steel Strengthening
Elyas Ghafoori
*Private report submitted to SBB: Ingenieurbro SIA HUBER+GEMPERLE, Schlussbericht Tragwerksbeurteilung Brcke I, Mnchenstein, Page 6, 2010.
Symm
45 m
Bay No.
1
FE Modeling
4.79
5.095
6 m 5.0
5.095
CL
5.095
5.095
5.095
Bay 2
Bay 1
5.095
Bay 3
5.095
Bay 4
4.79
Bay 5
6.15
24
26
5.0
Section A
Steel Strengthening
180/10
(x2)
90/130/10
(x4)
20
310
350
CL
160/9
(x2)
290
80/120/10
(x4)
140/7
(x2)
80/100/10
(x4)
70/70/8
180/8 (x2)
(x2)
Elyas Ghafoori
70/100/10
(x4)
140/7
(x2)
50/80/8
(x4)
140/7
(x2)
Stringer
PUR System
Prestressed un-bonded retrofit (PUR)
Cross beam
Stringer
Cross beam
1. Applicable to unsmooth surfaces (i.g.,

riveted beams).
2. Fast installation (no gluing & no surface
preparation).
3. Easy to prestress (no hydraulic jacks).
4. No traffic interruptions for bond curing.
5. Minimum damage (no hole, glue & grinding).
6. Adjustable prestressing level (to compensate relaxation).
7. Easy to remove.
Clamp (permanent)
Rivet
Cross beam
Prestressing chair (temporarily)
Column (permanent)
Cross beam
Steel Strengthening
Fibre Composites, 44
FS15
Elyas Ghafoori
Fatigue Theory
(MPa)
(MPa)
Before strengthening
After strengthening
Strengthening
Sa
Sm
Fatigue life
Material
properties
Stress
range
Mean
stress level
Not much increase in

stiffness
Steel Strengthening
Sm=(max+min)/2
Sa=(max-min)/2
Sy=yield limit
Su=ultimate strength
Se= fatigue endurance limit
Sa (Alternating stress)
Sy
Se
SAFE
-Sy Compression
Tension
Sy
Su
Sm (Mean stress)
Modified Goodman Constant Life Diagram (CLD)

Elyas Ghafoori
45
Laboratory Experiments
Cyclic loading
5000
Dimensions in mm
23
Detail of rivet holes in bottom

flange of beam
Steel Strengthening
Elyas Ghafoori
Steel Strengthening
Elyas Ghafoori
Sa (MPa)
417
270
1
45
Mode-I Fatigue Crack

Gerber line
Modified Goodman line

Sm (MPa)
-417
417
540
Fatigue test 1 (beam 1), not strengthened: F=[2.5-68] kN -> N= 500000 cycles -> cracked
Fatigue test 2 (beam 2), 30% prestressing: F=[2.5-68] kN -> N=2000000 cycles -> No crack
Fatigue test 5 (beam 2), 4% prestressing: F=[2.5-68] kN -> N=1500000 cycles -> cracked
Steel Strengthening
Elyas Ghafoori
Symm
Bay No.
1
4.79
5.095
5.0
5.095
CL
5.095
5.095
5.095
Bay 2
Bay 1
Bay 3
Adjustable column
5.0
Section A
4.79
Bay 5
180/10
(x2)
90/130/10
(x4)
20
310
160/9
(x2)
290
80/120/10
(x4)
3 CFRP plates
Bay 4
24
26
350
CL
Steel Strengthening
5.095
Magnetic strain
gauge
6.15
Mechanical friction clamp
5.095
Elyas Ghafoori
140/7
(x2)
80/100/10
(x4)
70/100/10
70/70/8
(x4)
180/8 (x2)
140/7
(x2)
(x2)
50/80/8
(x4)
140/7
(x2)
Symm
Bay No.
1
Bridge Strengthening
4.79
5.095
5.0
5.095
CL
5.095
5.095
5.095
Bay 2
Bay 1
5.095
Bay 3
Bay 4
5.095
4.79
Bay 5
6.15
24
26
Humidity and 350

90/130/10
temperature
sensors
(x4)
5.0
Section A
180/10
(x2)
Wireless sensor node
Steel Strengthening
20
310
CL
160/9
(x2)
290
80/120/10
(x4)
Elyas Ghafoori
140/7
(x2)
80/100/10
(x4)
70/100/10
70/70/8
(x4)
180/8 (x2)
140/7
(x2)
(x2)
50/80/8
(x4)
140/7
(x2)
5.095
4.79
Alternating stress,
5.095
a(MPa)
35
211.5CL
210
180
250
500
750
1000
1250
1500
1750
2000
82.5
0
-220 -150
Jumps due to
passenger train passages
Section A
-80
24
26
-10
60
180/10
(x2)
130
90/130/10
(x4)
200
160/9
(x2)
290
80/120/10
(x4)
270
140/7
(x2)
340
80/100/10
(x4)
70/100/10
50/80/8
70/70/8
(x4)
(x4)
180/8 (x2)
140/7
140/7
(x2)
(x2)
(x2)
Mean stress, m(MPa)
Time (S)
Stress-histories of CFRP plate and cross-beam bottom flange

while pre-stressing
CLD presentation of shifting stresses from finite life zone to infinite life
region using pre-stressed CFRP material (D4 load model)
Stress in Metal before Retrofit (MPa)

Stress in Metal after Retrofit (MPa)
CFRP Pre-stress Level (%)
60
40
20
-20
-40
0.5
1.5
2.5
3.5
4.5
5.5
6.5
Time (S)
Stress-histories of CFRP plate and cross-beam bottom flange

before & after strengthening due to S3 train
Steel Strengthening
Bay 5
20
310
350
CL
5.0
-35
Bay 4
4.79
106.1
90
30
-25
Bay 3
5.095
6.15
60
-15
Bay 2
5.095
150
120
-5 0
5.095
5.095
5.095
Bay 1
91.1
Stress in Metal (Mpa)

CFRP Pre-stress Level (%)
15
Modified Goodman
Modified Johnson
Yielding
60.2
25
Measurements
5.0
Bridge Strengthening
Measurements
A S3 passenger train crossing Mnchenstein Bridge
Elyas Ghafoori
Partners & Sponsors
Swiss Commission for Technology & Innovation (CTI)
Industrial Partners:
S&P clever reinforcement company AG
SBB (Swiss Federal Railways)
Research Partners:
EPFL, Steel Structures Laboratory
ETHZ, Structural Engineering Section
Steel Strengthening
Elyas Ghafoori
Conclusions: Prestressed vs. Non-prestressed CFRP Plates

Advantages:
Utilization of high tensile strength of CFRP materials

Prestressed FRPs can carry both a portion of the dead load and the additional live load carried by the
structure
Increasing yielding load

Increasing ultimate load capacity
Substantial increase in fatigue life
Possible arrest of existing fatigue cracks
Steel Strengthening
Elyas Ghafoori
Conclusions: Prestressed vs. Non-prestressed CFRP Plates

Disadvantages:
Large amount of labor work for prestressing (you can almost forget about CFRP cost).
High interfacial shear stresses at plate ends => earlier debonding => use of mechanical anchorage system
Steel Strengthening
Elyas Ghafoori
Conclusions: Bonded vs. Un-bonded CFRP-steel Composite Systems

Weak point:
The main difference between FRPsteel and FRPconcrete bonded joints is that in the former, failure will likely occur in the
adhesive layer and in the latter failure is expected to occur in the concrete. Thus the weakest point an FRP-steel composite
systems is the adhesive.
Surface preparation:
Prior to bond application, surface of steel beam should be cleaned and all paint and anti-corrosion coating have to be
removed.
High temperature:
Compared to concrete, steel has a high thermal conductivity (about 50 W/mK) and has significant ability to transfer heat
rapidly to the adhesive. Moreover, the rate of sunlight absorption by steel is much greater than the rate of steel
electromagnetic radiation (black body radiation); therefore, steel members exposed to direct sunlight on a hot day will easily
become much hotter than the ambient temperature. This effect makes the adhesive adjacent to a hot steel surface soften
excessively when the service temperature of the steel substrate approaches the glass transition temperature of the adhesive.
Metallic riveted bridges:

Due to the flat configuration of FRP plates, they cannot be bonded to the surface of structures that are not sufficiently smooth.
Because the cover plate is riveted to the steel girders in steel-riveted bridges, for example, there is a high rivet density and the
bonded FRP reinforcement system cannot be used.
Heritage structures:
The components of strengthening systems for heritage structures need to be designed for easy removal when there is a need to
restore the structure to its original unstrengthened construction design. In a bonded reinforcement system, FRP strengthening
materials cannot be easily separated from the beam due to the applied glue.
Steel Strengthening
Elyas Ghafoori
More details in: Ghafoori E, et al., Fatigue strengthening of damaged steel beams using unbonded and bonded prestressed CFRP plates, International Journal of Fatigue, 2012, 44 , pp. 303-315.
New Projects and Project Proposals
Steel Strengthening
Elyas Ghafoori
Strengthening of Aabach Bridge (90 years old ), Switzerland

Project partners:
SBB
dsp engineering office
Humidity and
temperature sensors
Wireless sensor node
The bridge is on the main line

between Zurich and Chur, just
outside of Lachen towards Chur.
Steel Strengthening
Elyas Ghafoori
Strengthening of Chandler Bridge in Melbourne, Australia

Chandler Bridge, a 130-year-old Bridge, in Melbourne, Australia, to be retrofitted by PUR system in 2016-17.
Steel Strengthening
Elyas Ghafoori
Interested in Doing Your Master Thesis at Empa?

The following projects are available at Empa for a semester or master thesis work:
Master thesis proposal:
1. Fatigue Crack Arrest in Steel Members using Pre-stressed CFRP Composites; a Fracture Mechanic Approach
This project aims to develop a model that can predict the required amount of CFRP pre-stress level to stop fatigue crack propagation.
The method is useful for strengthening of cracked metallic members such as bridge components and in aircraft industries.
2. Bond Behavior of Prestressed CFRP Laminates to Steel Substrate: Analytical Modeling and Experimental Verifications
The main goal of this master project is to investigate the effect of pre-stressing on bond behavior and residual bond capacity of CFRP
laminates bonded to steel substrate. The outcome of the project would be an analytical model (accounting for the pre-stressing level)
which shall be verified by experimental tests.
Semester project proposal:
3. Determination of the most fatigue prone details of Aabach Railway Bridge in Lachen using SIA codes
The project aims to study the documents of Aabach Steel Railway Bridge (near Lachen) and to verify the static and fatigue
performance of the bridge based on SIA codes. The project is in collaboration with SBB and dsp engineering office.
For more information contact: Elyas Ghafoori: elyas.ghafoori@empa.ch
Steel Strengthening
Elyas Ghafoori
References
Ghafoori E., Motavalli M., Zhao X.L., Nussbaumer A., Fontana M. Fatigue design criteria for strengthening metallic beams with
bonded CFRP plates. Engineering Structures, 2015. 101: p. 542-557.
Ghafoori E., Motavalli M., Nussbaumer A., Herwig A., Prinz G., Fontana M. Determination of minimum CFRP pre-stress levels for
fatigue crack prevention in retrofitted metallic beams. Engineering Structures, 2015. 84: p. 2941.
Ghafoori E., Motavalli M., Nussbaumer A., Herwig A., Prinz G.S., Fontana M. Design criterion for fatigue strengthening of riveted
beams in a 120-year-old railway metallic bridge using pre-stressed CFRP plates. Composites Part B, 2015. 68: p. 1-13.
Ghafoori E., Motavalli M. Normal, high and ultra-high modulus CFRP laminates for bonded and un-bonded strengthening of steel
beams. Materials and Design, 2015. 67: p. 232243.
Ghafoori E., Motavalli M. Lateral-torsional buckling of steel I-beams retrofitted by bonded and un-bonded CFRP laminates with
different pre-stress levels: experimental and numerical study. Construction and Building Materials, 2015. 76: p. 194206.
Ghafoori E., Prinz G.S., Mayor E., Nussbaumer A., Motavalli M., Herwig A., Fontana M. Finite element analysis for fatigue damage
reduction in metallic riveted bridges using pre-stressed CFRP plates. Polymers, 2014. 6(4): p. 1096-1118.
Ghafoori E., Motavalli M. Flexural and interfacial behavior of metallic beams strengthened by prestressed bonded plates. Composite
Structures, 2013. 101: p. 22-34.
Ghafoori E., Schumacher A., Motavalli M. Fatigue behavior of notched steel beams reinforced with bonded CFRP plates:
Determination of prestressing level for crack arrest. Engineering Structures, 2012. 45: p. 270-283.
Ghafoori E., Motavalli M., Botsis J., Herwig A., Galli M. Fatigue strengthening of damaged metallic beams using prestressed
unbonded and bonded CFRP plates. International Journal of Fatigue, 2012. 44: p. 303-315.
Ghafoori E., Motavalli M., Flexural and interfacial behavior of metallic beams strengthened by prestressed bonded plates, Composite
Structures, 101 (2013), 22-34.
More documents and articles in this field can be found in: www.researchgate.net/profile/Elyas_Ghafoori/publications
Steel Strengthening
Elyas Ghafoori
Thank You for Your Attention!

Any Questions?
Steel Strengthening
Elyas Ghafoori

Steel Strengthening - Elyas

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Steel Strengthening - Elyas

Enviado por

Direitos autorais:

Formatos disponíveis

Table of content of printouts:

Fibre Composites, FS15

Strengthening of Metallic Members using CFRP Laminates

Fibre Composites, FS15

Introduction & Motivation

Fibre Composites, FS15

Problems in Metallic Bridges:

Most commonly used structural metals:

Fibre Composites, FS15

Kuehn et al. Assessment of Existing Steel Structures: Recommendations for Estimation of

Kuehn et al. Assessment of Existing Steel Structures: Recommendations for Estimation of

Fibre Composites, FS15

Introduction: Why CFRP Laminates?

Classifications of the CFRP laminates according to their

Low modulus (LM)

< 100 GPa (ECFRP < 0.5 Esteel)

Normal modulus (NM)

100 200 GPa (0.5Esteel ECFRP < Esteel)

High modulus (HM)

200400 GPa (Esteel ECFRP < 2Esteel)

Ultra-high modulus (UHM)

400 GPa (ECFRP 2 Esteel)

Fibre Composites, FS15

Introduction: Change in CFRP Youngs Modulus

The measured Youngs modulus of the NM, HM, and UHM

Young's modulus, ECFRP [GPa]

Fibre Composites, FS15

Introduction: Some of Typical Failure Modes of Steel Beams

Fibre Composites, FS15

Introduction: Possible Failure Modes of CFRP-toConcrete/Steel Bonded Joints

Fibre Composites, FS15

Fibre Composites, FS15

Steel Beam Strengthened by a Prestressed Bonded Plate

Fibre Composites, FS15

(2) & (3) into (6) =>

Force equilibrium in x direction:

Fibre Composites, FS15

Beam strengthened by the bonded CFRP laminate in a four-point bending set-up.

From (1) & (7) =>

Stress in bottom flange:

Fibre Composites, FS15

A simply supported beam with a free span of L=1200 mm and a

Shear & Bending

Strain in CFRP (micro-strain)

Strain in CFRP (micro-strain)

Fibre Composites, FS15

Bonded CFRP laminate

Experiment: 0% pre-stress in NM CFRP

Force in actuator (kN)

Strain in CFRP (micro-strain)

The calculated and measured strains along the CFRP

The strain in the bottom flange of the specimens strengthened by the

Fibre Composites, FS15

Bonded CFRP laminate

B1: control (unstrengthened) beam

B4: bonded with the plate @ 40% prestress

Presented analytical method for B4

Comparison of load-deflection behaviors of steel beams with y=355

B3: control (unstrengthened) beam

Fibre Composites, FS15

Comparison of load-deflection behaviors of two steel beams

Strengthening against Lateral Torsional Buckling

Fibre Composites, FS15

Definition of Lateral Torsional Buckling (LTB)