Академический Документы
Профессиональный Документы
Культура Документы
Key Words:
seismic capacity design, reinforced concrete column,
failure modes, deformational ductility, flexural failure, shear failure,
shear failure after flexural yielding, shear-degrading curve,
nonlinear dynamic analysis, random deformation history
1
Dept. of Civil Engrg., Musashi Inst. of Tech., Tokyo, Japan.
2
Inst. of Tech., Tokyu Construction Co., Sagamihara-City, Japan.
Ductility and Failure Modes
of Single Reinforced Concrete Columns
Abstract
The main purpose of the seismic analysis of structures is to assess the level of risk
associated with loss of serviceability, restorability and collapse. With regard to a reinforced
concrete bridge column, it is extremely important to identify failure modes and to qualify the
deformational ductility from the point of the capacity design methodology (Paulay and
Priestley (1992)).
In this paper, types of failures of single reinforced concrete columns are classified as
flexural failure, shear failure after yielding of longitudinal reinforcement, and shear failure.
Procedures to determine failure modes are presented by comparing P − δ curve (push-over
behavior) and the degrading capacity of shear strength along the deformational history.
Nonlinear dynamic analysis is also carried out for reinforced concrete columns
3
Dept. of Civil Engrg., Musashi Inst. of Tech., Tokyo, Japan.
4
Inst. of Tech., Tokyu Construction Co., Sagamihara-City, Japan.
subjected to the recorded time history of strong seismic actions. The shear-degrading curve
proposed by Priestley et al. (1996) is extended to a case of random deformation history to
cope with actual seismic excitation. The proposed method is compared with data of
experimental works and numerical simulations are carried out.
Introduction
The shear strength is, on the other hand, gradually reduced due to large
cyclic deformation beyond the displacement of yielding of the main reinforcement.
The shear strength are thus denoted as:
ductility
fcctor
1 μmu
displacement δ
(A) shear failure (B) shear failure after yilding (C) flexural failure
envelope curve
P (P- δ curve)
degradation of
shear strength curve
δy δ mu δy δmu
δy δmu
Now that the failure modes can be defined according to the intersecting
relationship between the P − δ curve and the degrading shear capacity envelope
(Priestley et al. (1996)). This is
A) Shear failure: Shear failure occurs before the main reinforcement yields
B) Shear failure after flexural yielding: Shear failure occurs after the main
reinforcement yielding
C) Flexural failure: The P − δ curve and the degrading shear capacity
envelope do not intersect till reaching the ultimate flexural point (δ = δ mu ).
Photo 1. Test Examples for Three Failure Modes (Hattori et al. 1998)
Static Nonlinear Analysis and Shear Strength
Deformational Analysis
δ pullout caused by pulling out the main reinforcement at the column base. Namely,
δ pullout caused by pulling out the main reinforcement was calculated by applying
The well-known modified truss analogy was applied for the calculation of
shear strength in this study. This means that shear strength is obtained by
summing shear strength by lateral reinforcement Vs and strength of concrete
shear resisting mechanisms Vc (for example, JSCE Specification (1996)). In this
paper, the following expressions are used to consider initial shear strength and
degrading shear strength separately.
initial shear strength :V y 0 = V s + V c0
(3-a)
(3-b)
Vs = Aw f wy z cotθ / s
(4)
It has been pointed out that the reduction of shear strength accompanying
cyclic excessive deformation is caused by degradation of the component of
concrete contribution V c 0 . The symbol V ck is used as the strength in the
degradation process. Here we introduce the degradation curve proposed by
Priestley et al. (1996) as shown in Figure 2.
In Figure 2, the left vertical axis represents the shear strength τ c , the right
vertical axis the shear degradation factor ζ obtained by dividing shear strength
Vck by the initial shear strength V c 0 . The initial value of this degradation factor
ζ is equal to 1, and ζ < 1 for the larger deformation which is expressed such
that ζ = ζ ( µ d ) . Thus, the relationship between the initial shear strength τ c 0 and
the degrading shear strength τck as well as between Vc 0 and Vck are simply
written as
Vck = ζV c0 , τ ck = ζτ c 0 (5)
2.0
1.2
1.5 1.0
μd= 2 → ζ= 1.000
τ c(N /mm 2 )
0.8
μd= 4 → ζ= 0.345
ζ
1.0 μd= 8 → ζ= 0.172 0.6
0.4
0.5
0.2
0.0 0.0
0 2 4 6 8 10
displacement response factor
μd
column base. This proposed formula reflects the new theoretical consideration to
be proven by a wide range of experimental database).
Comparison with Test Results and Numerical Simulation
The analytical result of specimen C05 by the proposed method shows that
deformational behavior is quite similar to the test result and predicted the shear
failure identical to the test. The analytical results of specimen C10 using two shear
strength degradation curves intersected nearly at the same points on P − δ
envelop curve, and coincided with the failure mode of the test results (shear
failure after flexural yielding). The analysis of specimen C20 shows that both of
the shear degrading curves do not intersect with P − δ curve and the failure mode
is assessed to be of the flexural failure.
As an example the case for specimen C10 that failed in shear after yielding
of longitudinal reinforcement, is shown in Figure 3, where the test result and the
(a)Experimental (b)Numerical
200 200
lateral force:P(kN)
150 150
lateral force:P(kN)
100 100
50 50
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
displacement:δ(mm) displacement:δ(mm)
Numerical Simulation
It is known that the failure mode and ductility of a member are affected by
the ratio of flexural yield strength to shear strength, concrete contribution of shear
strength, main reinforcement ratio, lateral reinforcement ratio, axial force level,
and so on. Then parametric simulations are carried out based on this proposed
technique for those affecting factors. The shape of column used is identical to the
above specimens ( A =320 mm×320 mm and shear span ratio a =4.05). The
New-RC Guideline equation was applied to calculate the shear strength-degrading
curve.
Figure 4 shows the member ductility ratio vs. the ratio of shear strength to
flexural strength with parametric range used in this analysis. Here we denote the
ratio of shear strength to flexural strength as V y 0 / Vmu (flexural capacity Vmu of
From this figure, it may be suggested that each failure mode can be
approximately estimated by the ratio of shear strength to flexural strength in such
a way that
8 8
factor
μ:or
6 6
tyct
4
s・f f. r du ct ilifa
4
s・f . r ductility
2 2
s y atio
o
sy/ ati
0.2 0.2
c
f'c
/f'
0.4 0.4
: p reinf
: p rein
0.08 0.08 0.6
shear 0.06 0.6 shear 0.06
reinf. reinf. 0.04
in
0.04 in 0.8
: pw・fw ratio 0.8 : pw・fw ratio 0.02
ma
0.02
ma
y/f'c y/f'c
0.08 0.08
shear reinf. ratio : pw・fwy/f'c
0.04 0.04
0.03 0.03
0.02 0.02
0.01 0.01
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
main reinf. ratio : ps・fsy/f'c main reinf. ratio : ps・fsy/f'c
(a)σo/f’c=0 (b) σ
o/f’c=0.2
Figure 5. Evaluation of Failure Modes and Ductility Ratios for Two Axial Level
in Relation of Main/Shear Reinforcement Ratio
(a)
column base
+φy
curvature at
time
−φy
shear failure
Vy0
degrading shear
Vy1 strength :Vyk (t)
Vy2
lateral force
(b)
+δm u
(c)
displacement
column top
−δm u
Figure 6. (a) Curvature at Column Base, (b) Lateral Force Acting on Column Base
and (c) Lateral Displacement at Column Top in Time History
Response
Therefore the failure modes for the dynamic random response can be
classified into the same manner as previously defined in addition to the
non- failure. Determination of dynamic failure modes and maximum response
displacement µ max can be summarized as follows:
µ max < 1
B) Shear failure after yielding of longitudinal reinforcement:
V (t ) > V yk ( t ) →
µmu = µd
µd < 2 : ξ =1
2 ≤ µd < 4 : ξ = −0.3275mµd + 0.655m + 1
4 ≤ µ d < 8 : ξ = − 0 .04325 mµ d − 0.518 m + 1
8 ≤ µd : ξ = − 0.828 mµ d + 1 (7)
k
ζk = ξ1ξ2ξ3 ...ξk = ∏ ξi
i =1
(8)
μ
μ
μ ζ=ζ(μ)
+4δy
+4
+3δy
+2δy
(a)
0
time 0 ζ 1
-2
-4
+4 1
(b) μ
μ m
i=1 3 ξ
0 3
1 1
-2 2
2
-4 i=1,2,3・・・,k:sequential peak number
ζ
time 0 1
ζ,ξ
shear
flexural
failure modes non failure failure after
failure
yielding
Fig. Fig.8(a) Fig.8(b) Fig.8(c)
μmax 4.31 3.94 2.68
strength to flexural strength Vy 0 / Vmu are getting small in order of P1, P3 and P5
(see TABLE 1), and then the failure mode shifted from non-failure ( µmax = 4.31),
flexural failure ( µ max = 3.94 ), and shear failure ( µ max = 2 .67 ).
μ μ
μ
μmu =4.72 μmu =3.15
μmu =3.94
6 flexural failure
4 no intersect
displacementμd
2 t=6.50sec
0
-2
-4
-6
no intersect
degradation of degradation of
15 shear strength Vyk (t) shear strength Vyk (t)
10
lateral force V
5 t=6.52sec
V( V(t) V(
V(
0
-5
-10
-15
TIME(sec)
no intersect
TIME(sec)
V(t)
Figure 10 shows the degrading shear strength τck taking the maximum
ground acceleration as a parameter. It was found that, as the maximum input
acceleration ( α max : indicated in the figure) increases, the column causes the more
damage to become the lower shear strength τ ck .
(a) (b)
5
i=1
ζ
4 ξ JMA-KOBE(P3, αmax =800Gal )
2
3
μd
3
2
1
0
1 0.5 0
ζ,ξ
0 5 10 15 20
time(sec)
500Gal
1.5
600 700
τ c (N /mm 2 )
ξ
1.0
800 ζ
0.5
0.0
0 2 4 6 8 10
μd
Conclusions
Failure modes for a single reinforced concrete column were classified into
three types: shear failure, shear failure after flexural yielding and flexural failure.
It is especially difficult to model the shear failure after flexural yielding which has
been discussed from the viewpoint of seismic design procedure. The modified
truss analogy incorporating concrete degrading models proposed by Priestley and
New RC equation from AIJ Design Guidelines were utilized to predict the
degrading process up to the shear failure.
When maximum input acceleration was, for instance, α max = 800 Gal,
degrading strength τck got lowered to 0.6 - 0.8 N/mm2 whereas initial shear
strength τ co of concrete was 1.5 N/mm2 . On the other hand, the shear strength
τ ck for seismic design in the current Japanese specifications (JRA Specification
and JSCE Seismic Code) is approximately 0.3 - 0.5 N/mm2, which is found to be a
more conservative value.
References
An, X. and Maekawa, K. (1998). “Shear Resistance and Ductility of RC Columns after Yield
of Main Reinforcement.” Journal of Materials, Concrete Structures and Pavements, JSCE,
No.585/V-38, 233-247.
Architectural Institute of Japan (1990). Design guidelines for earthquake resistant reinforced
concrete buildings based on ultimate strength concept, Japan. (In Japanese)
Hattori, H., Miyagi, T., Masuda, Y., Iketani, K. and Yoshikawa, H. (1998). “Evaluation of
Failure Mode and Ductility of Reinforced Concrete Columns.” The 10th Japan Earthquake
Engineering Symposium, Proceedings Vol. 2, 2157-2162. (In Japanese)
Japan Society of Civil Engineers (1996). Standard specification for concrete structures,
Seismic Design, JSCE. (In Japanese)
Paulay, T. and Priestley, M.J.N. (1992). Seismic Design of Reinforced Concrete and Masonry
Buildings, John Wiley & Sons.
Priestley, M.J.N., Seible, F., and Calvi. G.M. (1996). Seismic Design and Retrofit of Bridges,
John Wiley & Sons.