See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268011240

Seismic performance of stairs in the existing reinforced concrete building

Article · January 2008

CITATIONS READS

13 2,146

3 authors, including:

Edoardo Cosenza Gerardo Mario Verderame

University of Naples Federico II University of Naples Federico II
303 PUBLICATIONS   7,890 CITATIONS    262 PUBLICATIONS   4,246 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Facing a holistic and forward-looking approach for a problem-solving and integrated disaster risk governance: insights into seismic fragility methods View project

COMPDYN 2023 - Minisymposium 33: Insight in seismic performance of retrofitted buildings: feasibility and effectiveness of intervention in RC buildings View project

All content following this page was uploaded by Gerardo Mario Verderame on 20 May 2015.

The user has requested enhancement of the downloaded file.

 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

SEISMIC PERFORMANCE OF STAIRS IN THE EXISTING REINFORCED

CONCRETE BUILDING
Edoardo Cosenza1, Gerardo Mario Verderame2, Alessandra Zambrano3
1
Professor, Dept. of Structural Engineering, University of Naples “Federico II”, Naples, Italy
2
Assistant Professor, Dept. of Structural Engineering, University of Naples “Federico II”, Naples, Italy
3
Research Fellow, Dept. of Structural Engineering, University of Naples “Federico II”, Naples, Italy
Email: [email protected]; [email protected]; [email protected]

ABSTRACT :

This paper deals with the seismic performance of existing buildings and in particular on the moment resisting frame
structures that could have their critical and weak points in the stair members: columns and beams or slabs. The stair
increases structural strength and stiffness of a structure but attracting seismic forces it could fail into its short columns
or into the slabs due to high shear forces, into inclined beams supporting the steps a cause of high axial forces. The
structural solutions and design practice of stairs in gravity load designed structures are investigated to define their real
geometric definition and to understand their performance. Some numerical modal linear and non linear push-over
analyses are herein presented. A typical reinforced concrete building respecting the materials and design criteria of
the time is considered for the analyses. In particular two types of stairs are considered: the one with cantilever steps
constrained in inclined beams, and the stair composed of simply supported slabs. The modal analysis emphasizes the
different modal behavior considering the stairs. A non linear lumped plasticity models allow to perform non linear
pushover analysis that allow to identify the main failure mechanisms. Some numerical simulations give some
interesting results and offer some good features on the problems related to the mechanical and geometrical modeling
of the structural elements of the stair, and to the principle types of failure due to flexure, or shear.

KEYWORDS: Reinforced concrete; gravity loaded buildings; stairs; short column; shear failure.

1. INTRODUCTION

There exists a large number of reinforced concrete (RC) buildings that are gravity load designed and constructed in actual
seismic areas. Many of these structures were constructed in areas that are not considered seismic at the construction time or
although they were located in seismic areas at that time, the earlier codes did not include seismic provisions or may have
specified lower levels of seismic loads. Due to the high cost of replacement, many old structures are still in service far
beyond their design life. Besides, gravity load designed structures may perform in a no-ductile manner with dangerous
modes of failure. Before the 1980’s the design of the structure, both in seismic and in non seismic area, did not consider the
presence of the stair, although the stair offers an higher strength and stiffness influencing considerably the distribution of
seismic forces. It is well known that the stair could be a vulnerable part of the structure attracting the seismic action, in the
meanwhile its stiffness could preserve the structure from collapse if it was adequately designed and built. If the stair is not
well designed it can lead the structure to collapse, in particular if only gravity loads are considered into the design or the
reinforcement detailing is not adequate.
It is of particular interest the identification of the weakest and most vulnerable elements of the structure, the type of
collapse and damage considering the presence of the stairs. In particular, short columns, cranked beam and slab are
vulnerable elements of the structure during seismic excitations. Experimental researches and post-earthquake
reconnaissance have demonstrated that reinforced concrete columns with lightly and widely spaced transverse
reinforcement are vulnerable to shear failure during earthquake (Sezen & Moehle, 2004). In the last case, shear damage is
accompanied by a reduction in axial load capacity. Such types of failure can be brittle if the shear strength is reached when
the column is still in the elastic range or of limited ductility when it is reached after yielding.
In the present paper an overview on the most common stair typologies and on their design practice is presented. The
potential types of failure of each element and of the whole structure have been investigated performing static nonlinear
analyses.
2. DESIGN PRACTICE OF STAIRS IN EXISTING BUILDING
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

2.1. Stair classification

From the early beginning of the use of reinforced concrete (RC), different types of stair have been designed. According to
the literature, the existing stairs can now be classified into two main categories depending on the static behavior of the stair
steps: (i) stairs with steps performing as cantilever beam, (ii) stairs with simply supported steps.
In side these two categories three principal types of stairs can be distinguished:
• stair type A (see Figure 1) – The stair structure is composed of: 1) columns, at least four columns are located at the side of
the staircase (generally at its four corners), but in some cases they can be located internally to the substructure “stair”; 2)
beams that connecting the columns (storey beam and inter-storey beam); 3) beams supporting the flight steps (element
bs1-bs2-bs3 in Figure 1). In particular, storey and inter-storey landings are supported respectively by elements bs1 and
bs3, while steps are cantilever beams constraint into the inclined part of the beam bs2. Three types of beam
configurations can be distinguished, depending on the presence of bs1 and bs3.
• stair type B (see Figure 2) – The substructure “stair” is composed by: 1) columns (at least four); 2) beams connecting
the columns, storey beam and inter-storey beam; 3) the slab constraint at the beams at each storey and inter-storey. This
slab has two horizontal parts (s1-s3) and one inclined on the horizontal (s2). On this cranked element the steps are simply
supported, they are made contemporary or successively to the slab. The slab can be made of only reinforced concrete or
of brick and joist.
• stair type C – The staircase is composed by reinforced concrete walls, and the steps, having a cantilever behavior are
fully constraint in these RC walls.
Generally the stair type B and C are used worldwide, in Europe and USA, while the stair type A, with cranked beams, are
much more adopted in Europe (Tecnica y Pratica del Hormigon Armado, 1989, Reynolds and Steedman, 2002, Guerrin and
Lavaur, 1971, Berry, 1999, Manual published by the Institution of Structural Engineers,1985).

2.2. Design practice

According to the manual design criteria (Marrullier, 1910; Rosci, 1939; Santarella, 1953, 1957; Pagano, 1963; Migliacci,
1977) stair type A could be designed considering only gravity loads, any seismic actions could not be taken into
consideration. The permanent and live loads on the steps generate on the beam element (bs1-bs2-bs3) a torsional moment T
and a distributed load producing on the beam shear force V and bending moment M. Each flight step is designed modeling
it as a cantilever beam subjected to a distributed load.
storey landing
am
am be
be rey slab
o rey lon sto
st git
ud
p ina be
ste
column

lb am
column

ea
m
column

column
interstorey height
interstorey height

am interstorey landing m
be ea
rey yb
column

o re
t sto
column

bs
1 s1
bs2 bs s2
3 lon s3
git
ud
ina
column

storey
column

storey lb
ea
interstorey beam m
interstorey beam

Figure 1 – Stair type A. Figure 2 –Stair type B.

As it is explained in several dated manuals (Santarella, 1953; Pagano, 1963), the maximum M+ and minimum moments M-
into the beam (bs1-bs2-bs3) is evaluated on the basis of different static schemes corresponding to different constraints at
the extreme ends of the beam. Two extreme constraint conditions to evaluate the bending moment are suggested: (i) full
constraint at the extreme ends, (ii) simply supported, the extreme end of the beam can rotate. The effective bending
moment value is obviously in between the two solutions. The torsional moment is considered of relevant importance, it
leads to add transverse reinforcement (stirrups) along the length of the beam (bs1-bs2-bs3). The beam scheme is linear,
equal to the projection on the horizontal of the real beam. The adopted values of the torsional moment depend on the
hypothesis upon the flexural stiffness of the inter-storey slab: flexible and rigid diaphragm (see Figure 1b). The shear force
is generally considered to design the transversal reinforcement (stirrups); some manuals emphasize on the importance of a
correct reinforcement bar location into the beam section, no acute angle that could produce a pull out of the concrete cover
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

should be made on the reinforcement bars.

About stair type B, manuals indicate two limit structural schemes: (i) an horizontal beam full constraint at the end, (ii) an
horizontal beam simply supported at the extreme ends. Normally bending moment and shear are the internal forces taken
into consideration. In the books and manuals of the construction time the only severe prescription is regarding the design of
the steel bars: a reinforcing bar should not bent to form angles that favorite the pull out of the concrete cover (Pagano,
1963).
In literature few prescriptions can be found to design a structural RC wall (stair type C) at which are full constraint the stair
flight steps, although these walls attract seismic actions, just few notes regarding the effect of the horizontal action on the
moment distribution along the stair height, are remarked in literature (Migliacci, 1977). Each step is designed as a
cantilever beam, a small slab connects the separated steps.
The Italian stair design practice during the period 1954-1980 has been analyzed in order to identify the most common
typologies and the effective adopted design criteria. A sample of 60 buildings designed before 1980’s in Southern Italy in
no seismic areas and 30 buildings in seismic areas has been collected and analyzed. The predominant stair type in the
studied building sample is type A. On the basis of this statistical analysis on the sample a typical building is designed, and
the results of numerical simulation on the building model are reported in the following paragraphs.

3. SEISMIC VULNERABILITY OF STAIRS

The presence of stairs can influence the existing building resistance, this is due to: distribution of seismic forces that is not
considered in the design, the different modeling design of stair structure, material strengths, element detailing.
In general, whenever the presence of a stair in a building is considered, it highly influences the seismic response of the
generic reinforced concrete structure. The structural typology of stairs generally introduces discontinuities into the typical
regular reinforced concrete skeleton, composed of beams and columns. The considered sub-structure “stair” is an
assemblage of inclined cranked elements, as slabs or beams and columns (stair type A, B). All these elements contribute to
increase the stiffness of the stair, due to the static behavior of inclined elements and of short columns. For all these reasons
the elements that constitutes the stair are often characterized by a strong seismic demand. The short columns are subjected
to high shear demand that can lead to a premature brittle failure and to a limited rotation capacity of the element. The
cranked beams, differently from the horizontal beam, are defined by high variation in axial forces that can modify bending
capacity, resistance and deformability of all these elements.
On the other side, in the practice these elements (cranked beams and stair columns) have been designed for only gravity
loads. Short columns in the stairs have longitudinal and transversal reinforcements equal to the other columns of the
structure having normal height, with possible failure due to shear before or after yielding (brittle or limited ductile failure).
The cranked beams have longitudinal reinforcement with insufficient overlapping and poor detailing. The inter-storey
landings, between two successive cranked beams, are generally lightly transversally steel reinforced, so they result
vulnerable to shear. These considerations evidence that stairs are potential vulnerable points of structures subjected to
earthquakes, while the stair elements have low ductility capacity and may be subjected to possible brittle failures.
The non linear lumped-plasticity modeling of the stair elements is not easy and immediate to be correctly adopted, in the
following it is shown how different modeling could yield to different responses. Columns of a rectangular shape frame are
usually characterized by flexure in two distinct bending planes (parallel to both sides of the section), while beams are
defined by flexure in a single vertical bending plane. In fact, the adoption of a shell or a diaphragm constraint into the floor
modeling (diaphragm/shell) yield to negligible or absent bending moments in the beams constraint by the floor. On the
other side, elements of the stairs not belonging to the building floor (cranked beam and slabs, inter-storey beams, and
inter-storey landings) and not constraint by a diaphragm/shell would be modeled as columns having biaxial flexure.
All these aspects are discussed with more details in the following with a series of analysis on a RC building representative
of the studied sample. First, some linear analyses are conducted to evaluate the influence of the stair on the dynamical
characteristics of the building. Successively a series of non linear static analyses (static push over analysis) finalized to the
evaluation of the role of the stairs, of their elements and modeling is preformed. Two models have been considered to study
stair type A (with cranked beams) and stair type B having reinforced concrete slab. For each geometric structure, different
modeling have been adopted to evidence the influence on the global response of biaxial flexure modeling in the beams of
the substructure “stair”, bending moment-axial force (M-N) interaction into the cranked elements (beam and slabs),
bending moment-shear (M-V) interaction into the cranked elements and columns.

3. NUMERICAL ANALYSIS
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

3.1 The analyzed structures and its model

A typical building has been designed with the aim of performing numerical simulations and capturing its seismic
behavior with different types of stair: it has a rectangular shape, it is symmetric respect a transversal axis, its dimensions in
plant are 19.00×10.00m with three floors (first height 4.5m, inter-storey height 3.0m, total height above the foundation
10.50m). The structure is constituted by parallel frames (three in the longitudinal direction and two in transversal
direction), as shown in Figure 3.
column beam (30x50)cm
13 14 15 16 17 18 (2+ 2)φ16
stirrups φ6/10cm
30x30 30x30 30x30 30x30 30x30 30x30 2φ16 2φ16 4φ16
5.00

2φ16 2φ16 6 2φ16

150 cm
1
30x30 35x35 35x35 35x35 35x35 30x30
(2+ 2)φ16 2φ
stirrups φ6/10cm
7 8 9 10 11 12
stair 16 9-10
4φ16 4φ
5.00

stirrups φ6/20cm

30x30 30x30 30x30 30x30 30x30 30x30

1 2 3 4 5 6 2φ16 stirrups φ6/10cm

4.00 4.00 3.00 4.00 4.00

3-4

Figure 3 – Plan of the studied building Figure 4 – Detailing of a cracked beam

The building has a gravity loaded design. The columns’ dimensions vary from (0.30×0.30)m to (0.35×0.35)m, each floor
slab has an height of 0.22m. The longitudinal steel reinforcement of the columns is of 8φ12cm while the transverse steel
reinforcement is constituted by stirrups of φ6/25cm. It is possible to notice that the geometric transverse steel ratio does not
respect the codes of the period of construction; in fact, the minimum longitudinal stirrup spacing may be equal to half of the
column section dimension (s=15cm).
The beam cross section is (0.30×0.50)m. The longitudinal beams have the following reinforcement steel: 4φ16cm at the top
and 2φ16cm at the bottom of the section at the ends of the beam, while in the central part of the beam 2φ16cm at the top and
4φ16cm at the bottom of the section. The transverse beams are lightly reinforced, composed of 4φ12cm at the top and
2φ12cm at the bottom of the section, and the transverse reinforcement is composed of stirrups φ6/10cm. In Figure 4 the
dimensions and steel reinforcements of the cracked beams are reported. The typical material strength of the time of
construction are adopted: the concrete is defined by a compressive strength fc=20 MPa (R.D. 2229, 1939) and a steel
yielding strength fy=350 MPa.
Two types of stairs are considered (see figure 5):
A. stair type A (with inclined beam and cantilever steps), in this typology the studied stair substructures differ by the
presence of the following elements: 1) without inter-storey landings and longitudinal storey beams (stair model
#1); 2) with inter-storey landings modeled as beam elements (stair model #2); 3) with inter-storey landings
modeled as beam elements and longitudinal storey beams (stair model #3);
B. stair type B with each cranked reinforced concrete slab modeled as beam element and supported flight steps (stair
model # 4).
For all these typologies a structure without stair is considered as a referee (stair model #0 in figure 5).

model #0 model #1
model #2 model #3 Model # 4
without stair stair type A Stair type B
Figure 5 – Staircase modeling.
The structural modeling has been performed with SAP 2000 (SAP2000, 2002). The geometric-structural modeling has
been chosen in order to represent in detail the geometric characteristics of the structural elements and the distribution in
plane and in elevation of the structural masses. The presence of floors is modeled with diaphragm constraints. The model is
characterized by the total absence of interaction with substructural soil; at zero level the columns are connected to the soil
with a full constraint (restraint). The non linear behavior of a generic structural element is studied by using a
lumped-plasticity model. Flexural hinges at the extreme ends of each elements are characterized by a moment-rotation
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

relationship (M-θ) of type: elastic until cracking, Mcr; cracked until yielding (θy, My); plastic with ultimate rotation and
moment (θu, Mu=My).To account the shear behavior, shear hinge with elastic-brittle behavior is added in parallel to each
flexural hinge. The shear resistance is evaluated following the formulation of Sezen & Moehle (2004).

3.2. The numerical results

Modal dynamic analyses are conducted on the five models in order to understand the influence of each stair element on
the dynamical behavior of the whole structure. In Table 1 the modal analyses results are summarized, in particular the first
two translational modal periods are reported: in transversal direction Y and in longitudinal direction X. The presence of the
stair in model #1, 2, 3, 4 increases the transversal stiffness and the period in Y direction drastically decreases to values
T=0.668, 0.660, 0.648, 0.642 sec. respectively for the different models respect to the result of model #0 without stair
(T=0.958 sec). This confirms the stiffness increase due to the stair presence whatever typology is adopted (stair type A or
B). In particular, the comparison with results in transversal direction Y, shows that the substructure “stair” contributes for
about the 50% to whole building stiffness. In the longitudinal direction X the period decreases moderately from the one of
model #0, since the introduction of the cranked elements (beams or slabs) does not increase significantly the stiffness in
longitudinal direction. The comparison among models #1, 2, 3 shows that the presence of inter-storey landings and
longitudinal storey beams does not affect significantly the first modes.

Table 1 – Modal analysis considering the different stair modeling.

Building Period [sec]
direction without stair with stair
model #0 model #1 model #2 model #3 model #4
X – longitudinal 0.729 0.724 0.719 0.718 0.715
Y – transversal 0.958 0.668 0.660 0.648 0.642

The conventional static analysis has been performed to evidence the failure modes and the influence of modeling on the
structural response. In the push-over analysis a vertical distribution of lateral loads is chosen proportional to the horizontal
lateral force distribution obtained by a modal elastic analysis at the first translational mode.
The results of the non linear analysis, for both transversal Y and longitudinal direction X, are expressed in terms of base
shear versus top displacement of the building. Moreover, each analysis evidences the first failure in an element; the ductile
failure is defined by a rotational demand equal to the ultimate cord rotation given by Eurocode 8, while the brittle failure is
defined by a shear demand equal to shear strength given by the model of Sezen and Moehle (2004).
Figure 6 shows the results of the analyses conducted on the building without stair (model #0). Two different behaviors are
evidenced in the two directions. In particular, the building in the longitudinal direction presents a greater stiffness, this
confirms the preceding non linear results, with a global resistance greater of about 90% respect to the transversal
direction. The absence of shear failure yields to a greater displacement capacity, conventionally limited to a ductile failure
of the column elements, of about ∆=0.21m in longitudinal direction and ∆=0.35m in transversal direction.
Moreover it can be observed the effect of the bending moment-axial force interaction (modeling M1+N) in the global
response that is not appreciable both in longitudinal and transversal direction. Such a result is clearly due to a low flexural
resistance of the beams (designed for only gravity load), to a limited shear increment, to a negligible variation of axial
force and to the correspondent flexural resistance of the columns when subjected to horizontal seismic force distribution.
The presence of stair in model #1 influences significantly the building response (see Figure 7). The pushover curves in
transversal direction evidence a greater lateral stiffness, an increment of resistance Vb, in mean of about the 100%, a
reduction in displacement capacity (∆≤0.10m) comparing the results with the ones for model #0. Moreover, the interaction
between the flexural mechanism and the axial force is not negligible; the variation of the flexural resistance with the axial
force (modeling M1+N) yields to a reduction in resistance and displacement capacity, respect to a uniaxial flexural model
(M1). This result can be interpreted comparing the results with the ones obtained for the building without stair (model #0),
in which the interaction bending moment-axial force M-N does not produce any effect. The M-N interaction assumes a
different importance in the building with stair, this is justified by the presence of cranked elements where the behavior is
governed by the axial forces: an high variation of axial force yields to a consistent variation in flexural resistance.
Comparing the results with the ones of model #0, in the longitudinal direction a greater resistant capacity in model #1 is
also registered a cause of the presence of the stair. On the other side, the interaction between the flexural mechanism and
the axial force does not produce any effect.
Moreover, the analyses evidence two singular aspects: the hardening behavior of the base shear-top displacement capacity
curves shown in both directions and the influence of the stair on the longitudinal response of the building. In transversal
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

direction, the first aspect is justified by the axial forces into the cranked beams that increase with increasing imposed
displacement. This result is also confirmed by the different response registered with the simple flexural uniaxial modeling
(M1) respect to the flexural modeling with bending moment-axial force interaction (M1+N). On the other side, this
consideration does not justify the influence of stair in longitudinal direction and its hardening behavior. This result should
be interpreted by the presence of elastic resistance due to the stair elements. The presence of the inclined beam at the first
level cause an high dissymmetry in the deformed shape; consistent relative displacements into the inter-storey beams are
registered in the transversal direction as well as in the inclined beams in longitudinal direction. According with the
common hypotheses of uniaxial plastic modeling of the beams (modeling M1) the relative displacements cause high
elastic internal forces. Moreover, this focus on an important issue that it is strongly suggested to do not model the cranked
beams and inter-storey beams of the stair in uniaxial manner in a lumped-plasticity approach. The asymmetric behavior of
the stair, caused by the restraint in the first inclined beam justifies the adoption of a biaxial modeling (M2) of the stair
elements (not belonging to the building floor) as it is commonly done for columns. The adoption of a biaxial modelling
(M2) produces different results respect to the ones obtained with preceding analyses obtained by an uniaxial modelling
(M1). In Figure 8 the base shear-top displacement curves for the building with stair model #1 obtained with biaxial
modelling (M2) are reported. It is interesting to observe that the absence of some critical aspects that characterize the
results of the preceding analyses. First, the longitudinal direction is not influenced by the presence of cranked beams; the
building has a resistant capacity lightly greater (of about 10%) and a minor displacement capacity (of about15%) respect to
the case of stair’s absence (model #0). The greater resistant capacity is due to the presence of short columns, that a cause
of their reduced shear span have a greater shear demand, and to the flexural resistance offered by the first inclined beam.
On the contrary the minor displacement capacity is caused by a reduced rotation capacity of the short columns; in fact, the
ductile failure is reached exactly into the short columns of the first level. Besides, the base shear-top displacement response
curves do not evidence any hardening behavior, but according to the element flexural modeling it presents a resistant
capacity invariant with displacement (global collapse mechanism).
1500 1500
M1 Vb [kN] M1 Vb [kN]
M1+N ductile failure M1+N
1250 1250 ductile failure
M1+N+V brittle failure M1+N+V
brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

longitudinal direction transversal direction

Figure 6. Stair model #0. Results of the push-over analyses.

1500 1500
M1 Vb [kN] M1 Vb [kN]
M1+N ductile failure M1+N ductile failure
1250 1250
M1+N+V brittle failure M1+N+V brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

Longitudinal direction transversal direction

Figure 7. Stair model #1 – modeling M1. Results of push-over analyses.
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

1500 1500
M2 Vb [kN] M2 Vb [kN]
M2+N ductile failure M2+N ductile failure
1250 1250
M2+N+V brittle failure M2+N+V brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

longitudinal direction transversal direction

Figure 8. Stair model #1 – modeling M2. Results of push-over analyses.
1500 1500
M2 Vb [kN] M2 Vb [kN]
M2+N ductile failure M2+N ductile failure
1250 1250
M2+N+V brittle failure M2+N+V brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

longitudinal direction transversal direction

Figure 9. Stair model #2. Results of the push-over analyses.
1500 1500
M2 Vb [kN] M2 Vb [kN]
M2+N ductile failure M2+N ductile failure
1250 1250
M2+N+V brittle failure M2+N+V brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

longitudinal direction transversal direction

Figure 10. Stair model #3. Results of the push-over analyses.
1500 1500
M2 Vb [kN] M2 Vb [kN]
M2+N ductile failure M2+N ductile failure
1250 1250
M2+N+V brittle failure M2+N+V brittle failure
1000 1000

750 750

500 500

250 250
∆ [m] ∆ [m]
0 0
-0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30

longitudinal direction transversal direction

Figure 11. Stair model #4. Results of the push-over analyses.
Moreover, it is not registered any consistent effect due to the bending moment–axial force interaction. On the other side in
the transversal direction the push-over curves present an hardening behavior, although smooth and less evident respect to
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

the preceding uniaxial modeling (M1), caused by the increasing axial force that is predominant into the inclined beams.
Furthermore, the bending moment-axial force interaction is not negligible, in fact it causes quite an effect respect to
modeling M1 in Figure 7, increasing the capacity curve with displacement. This result is justified on one side by the
absence of elastic parasite reactions of the inter-storey floor and on the other side by the great influence of the bending
moment-axial force interaction only on the cranked beams. However, the presence of inclined beams and short columns
yield to a greater stiffness and resistant capacity (of order 70%), on the contrary to a minor displacement capacity. The
ductile failure due to flexure is reached into the short columns of the stair. It is remarkable that the presence of short
columns exactly influences the response of the building with stair. In fact, if the building without stair (model #0) has not
modeled considering any interaction between bending moment and shear (see Figure 6), the building with stair (model #1)
registers a response highly influenced by shear failure into short columns. In Figure 7 the results of the analysis conducted
accounting the bending moment-shear interaction (modeling M2+N+V) are reported, a reduction into the capacity is
registered, it is due to a brittle failure in both directions of seismic input. The greater shear demand into the short columns
respect the common height columns caused by the minor shear span and by the lightly transversal steel reinforcement
justifies the brittle failure of these elements. The addition of the inter-storey landings (model #2) and of the eventual
longitudinal storey beam (model #3) does not modify substantially the building response that is highly more characterized
by the presence of the cranked beams (model #1) (see figures 9 and 10). The building with reinforced concrete slab (model
#4) has a similar behavior to the ones of the other models, presented in the preceding, as it is shown in Figure 11. The
response differs in longitudinal direction in which some contradictory results are evidence if they are compared with the
ones of model #3. In the analyses with flexural modeling (M2, M2+N) the building presents a greater resistant capacity due
to a consistent contribution given by the flexural resistance of the first cranked slab. Besides, accounting for the bending
moment-shear interaction the analyses evidence a minor capacity caused by a premature failure due to shear into the
cranked slab at the first level, that precedes a shear failure into a short column of the stair.

CONCLUSIONS

The paper focuses the attention on the role of stairs in the seismic response of reinforced concrete buildings. A series of
linear modal and non linear analyses on a typical building has been preformed. With the aim of evidencing the influence of
all the elements constituting the substructure “stair” on the structural response of buildings, different stair typologies have
been analyzed: with steps constraint into inclined beams, steps simply supported by reinforced concrete slabs, and different
modeling have been adopted with increasing level of accuracy. In general, the presence of stair yields in the transversal
direction to an increase of strength and to a reduction in deformation capacity respect to the building without stair. On the
contrary, the results have confirmed the need to utilize biaxial bending modeling and to account for the interaction of the
different internal forces, such as bending moment-axial force interaction that characterizes the inclined elements, and the
bending moment-shear interaction that governs the behavior of short columns. In the studied case, as soon as more refined
modeling are used, shear failure becomes predominant in the short columns and in the reinforced concrete slabs and
precedes the conventional ductile failure due to pure flexure.

REFERENCES

Barry R., The construction of buildings (1999). 5th edition Vol. 2. Blackwell Science.
Guerrin A., Lavaur R.C.(1971), Traté de beton armè, Dunod (in French).
Manual for the design of reinforced concrete building structures. (1985) , by the Institution of Structural Engineers.
Marrullier E. (1910), Guida Pratica per la costruzione degli edifici con speciale riguardo al cemento armato, Ed. Torinesi
(in Italian).
Migliacci A. (1977), Progetti di strutture, raccolta delle lezioni tenute presso il Politecnico di Milano negli anni accademici
1966-67 1967-68, II, Ed. Masson, Milano (in Italian).
Pagano M. (1963), Strutture, Liguori (in Italian).
Reynolds C. E., Steeedman J.C., (2002). Reinforced concrete designer’s handbook, Spon Press Taylor & Francis Group.
Rosci L. (1939), Manuale pratico di volgarizzazione del Calcolo del cemento Armato, Lavagnolo, II, Torino. (in Italian).
Santarella L. (1953), Il cemento armato, le applicazioni alle costruzioni civili ed industriali, II Ed. Hoepli. (in Italian).
SAP2000 – ver.10 (2002), Computers and Structures Inc., Berkeley, California.
Sezen, H. & Moehle, J.P. (2004), Shear Strength for lightly reinforced concrete column, ASCE Journal of Structural
Engineering, 139:11, 1692-1703.
Tecnica y Pratica del Hormigon Armado. (1989) Ceac ed. Barcelona (in Spanish).

View publication stats