THE SEISMIC ASSESSMENT OF SOFT-STOREY BUILDINGS IN
MELBOURNE
JSITTIPOOM RODSIN, JOHNWILSON, NELSON LAM AND HELENGOLDSWORTHY
THE UNIVERSITY OF MELBOURNE
ttipoom Rodsin is a post-graduate student in the Department of Civil and Environmental
gineenng at The University of Melbourne
elson Lam is a senior lecturer in the Department of Civil and Environmental Engineering at The
niversity of Melbourne.
ohn Wilson is an associate professor in the Department of Civil and Environmental Engineering at
The University of Melbourne.
Helen Goldsworthy is a senior lecturer in the Department of Civil and Environmental Engineering at
The University of Melbourne.
ABSTRACT:
Buildings with soft-storeys are well known to be particularly vulnerable to collapse and
severe damage under earthquake excitation. Despite this, buildings possessing soft-storey
features are commonly found in low to moderate seismic countries such as Australia. Cost
effective retrofitting of soft-storey buildings to improve seismic performance is possible if the
seismic demand and building capacity can be realistically estimated. Conventional seismic
design procedures involve the determination of the building's seismic base shear demand
using a structural response factor to account for inelastic action and overstrength. Such
methods are approximate and do not provide guidance on likely seismic performance under
an extreme event.
This paper presents a simple and effective non-linear static procedure to compare the seismic
demand and capacity of a soft-storey structure. The seismic demand is represented by a
recently developed response spectrum model that realistically estimates the maximum
acceleration, velocity and displacement response in ADRS format. The capacity of the soft-
storey is ascertained using a deformation model that accounts for the effects of axial
compression, flexure, shear, column-end rotation and foundation flexibility. The uneven
sharing of shear forces between the columns and the significant additional shear demand
associated with strut actions in masonry infill are amongst the numerous important issues that
are raised for special attention. A proposed experimental program to evaluate the accuracy of
the analytical procedure is described.
Paper No. 17
1. INTRODUCTION
A soft-storey possesses horizontal stiffness and/or strength properties that are much less
than that of adjacent storeys and are typically the weak link in the resistance of the
building to horizontal loading. The concentration of damage in the soft-storey often
results in building collapse under severe seismic excitations. Soft-storey construction is
commonly associated with poor seismic performance and is prohibited by regulatory
controls in most high seismic regions. However, buildings featuring soft-storeys are still
commonly found in regions of low or moderate seismicity including Australia. A recent
survey by the authors of high-rise apartment buildings around the metropolitan area of
Melbourne revealed the widespread use of reinforced concrete portal frames, which are
designed to transfer vertical and horizontal loads from the first floor level to the ground
floor level. The flexible portal frames which constitute a soft-storey are designed to
support the "rigid-block" like upper floors consisting typically of reinforced concrete
walls and slabs. This paper presents the findings from an analytical investigation into
the seismic response behaviour of typical soft-storey buildings using a case study
building.
The traditional force-based approach in current codes of practices uses a "structural
response factor" (Rf) to account for both overstrength and inelastic response behaviour
of a building structure. The complex non-linear behaviour of a soft-storey is influenced
significantly by factors such as axial compression and shear span-depth ratio of the
columns and cannot possibly be modeled accurately and reliably by a single Rf
parameter. Furthermore, there are no specific provisions in current codes of practices
(e.g. AS1 170.4, 1993) to address soft-storey structures.
Non-linear static procedures (NSP) seem to be the most suitable method for assessing
the seismic performance of buildings possessing soft-storey features. In this paper,
results from analyses using NSP are presented in the acceleration-displacementresponse
spectrum (ADRS) format [ A T C a , 19961 to illustrate the performance of a soft-storey
building.
2. CAPACITY PREDICTION OF SOFT-STOREY BUILDINGS
The force-displacement behaviour of a typical soft-storey building is illustrated in this
section using a case study example. The lateral displacement of the soft-storey primarily
results from the deformation of the columns consisting of flexural deformation, shear
deformation, yield penetration and end joint rotation.
Flexural deformation can be relatively accurately estimated by integrating curvatures
that have been calculated in accordance with representative stress-strain relationships of
both the concrete and steel [Watson et a]., 19941 assuming plane section remaining
plane. Shear deformation 1s particularly significant with short columns possessing low
shear-span to depth ratios. In this study, a truss analogy method developed for cracked
concrete [Park and Paulay, 19751 was used to predict shear deformation assuming linear
elastic behaviour of the concrete "struts" and the steel "ties". An alternative method
would be to use compression field theory [Vecchio and Collin, 19861.
Page 17-1
The effects of yield penetration in the column longitudinal reinforcement at the
anchorage to the foundation was calculated in accordance with the recommendations by
Alsiwat and Saatcioglu [1992]. Finally, end rotations of the column contributed mainly
by the flexibility of the piled foundation and the connecting ground beams have been
incorporated into the analysis according to the recommendation in ATC4O [1996]. The
relative contributions from each of the deformation mechanisms to the total deformation
of the column have been studied by push-over analysis using the example column
shown in Figure 1.
The displacement behaviour of the portal frame as a whole was also studied. Analysis
results show that compressive stresses in the columns arising from the "push-pull"
actions associated with the application of the horizontal load could be very significant
(refer Figure 2). The initial axial load ratio of 0.15 under gravity loading could be
increased to 0.3 as the horizontal force is applied to the frame. Importantly, the stiffness
properties of the column are very sensitive to the induced axial compression.
Consequently, columns within the same portal frame possess very different stiffhess
properties. The force-displacement relationships (defining the effective stiffness) of the
individual columns along with that representing the portal frame as a whole are
presented in Figure 3 (points 1 to 3). The analytical deformation model described
previously (refer Figure 1) was augmented by the empirical model of Panagiotakos and
Fardis [2001] to extrapolate the response to failure. (i.e. beyond the point annotated with
a "3" in Figure 3)
The differential stiffness in the columns resulted in a very uneven sharing of the
horizontal shear forces between the columns within the portal frame. In the presented
case study of a 13-storey building (Figure 2), the more heavily loaded column (i.e.
column subjected to a higher compressive stress) attracted twice the shear force of a
lightly loaded column. Such differential load-sharing between the columns is typically
not modeled by the commonly used structural analysis packages. Thus, the seismically
induced shear forces in soft-storey columns are often understated by conventional
analyses.
Finally, the calculated displacements of the superstructure (associated with tilting of the
foundation) and P-delta effects have been included in the analysis to obtain the force-
displacement relationship of the whole building (refer Figure 3).
Page 17-2
-
-
~ o t deformation
d include end rotation)
~ l e x + a l +shear deformation + yield penetra$o{
:
----
- - - - Fl0?1^11?1-+.^031'^^1:"?-"0" -
I 6 reinforcement 1.5 6 reinforcement
Section A-A Section B-B
410 x 785 mm 410 x 630 mm
Height of Column (m)
Figure 1. Deformation along the height of column subjected to horizontal shear
Seismic
Left Column Right Column
Figure 2. Portal frame under axial load variations
Page 17-3
900
800
-& 700
800
500
a
2 400
0
"- 300
200
100
0
0 20 40 60 80 100 120 140
Displacement (mrn)
Figure 3. Shear - displacement behavior of columns, portal frame and whole building
3. PERFORMANCE EVALUATION OF SOFT-STOREY BUILDINGS
The potential seismic performance of the soft-storey building was assessed by the
capacity spectrum method (CSM) [ATC-40, 19961 based on the force-displacement
relationships developed in Section 2. An introduction to CSM can be found in Wilson
and Lam [2003]. The assumed seismic demand was based on provisions incorporated in
the draft Joint Australianmew Zealand Standard for Earthquake Actions [ASmZS
1170.4 Draft No.8: 20031. Information from borehole records taken from the area
indicates a Class C site. The performance point obtained by intercepting the demand
curves with the force-displacement (capacity) curves has displacement in the order of
20mm and the corresponding acceleration is 0.14g approximately (refer Figure 4). This
may conclude that under the earthquake, this building will perform satisfactorily under
Parameter in Draft 8 SANG
4-Â¥
,, ,
Z=0.08
Class C sits: F. = 1.2, Fv=1.5
...
.... . . .. .;. . .I Return period = 500 years
RSD Im)
Figure 4. Capacity spectrum diagram of the case study building
Page 17-4
4. ASSESSMENT OF ELEMENT PERFORMANCE
The shear capacities of the columns and possible "weak links" including beam-column
joint failure have also been checked against the shear demand. In calculating the shear
demand, the effects of the masonry infill have also been taken into account.
The shear capacity of the columns was estimated with reference to a multitude of code
provisions including that of AS3600 [2001] and ACI-318 [2002] along with the well-
known recommendations by Priestley et al. [I9941 and Moehle et al. [2002]. The 500-
year return period shear demands on the columns without masonry infill were found to
be significantly less than the shear capacity.
The introduction of concrete cavity masonry infill wall panels between RC columns at
ground level was found at a limited number of locations. The stiffening effect of the
infill wall can be idealized as a diagonal strut that braces the portal frame until the wall
strength is reached. This strut action could induce excessive local shear forces in the
adjacent RC columns and result in the shear resistance of a few of the ground floor
columns being exceeded. This was not considered critical to the. overall stability of the
building since alternative load paths exist in adjacent portal frames and the global
response of the structure is displacement and not force controlled. In view of the limited
displacement demand of the earthquake (20mm), the failure of a column in shear would
not lead to the collapse of the building provided that the column is effectively braced by
adjacent lateral load resisting elements, which perform satisfactory.
The shear capacity of the beam-column joints has also been assessed in accordance with
the estimates of the forces that could be transmined from the adjoining beams and
columns. The resulting stress conditions associated with "joint opening" and "joint
closing" actions have been analysed using the Mohr's circle approach based on
recommendations by Priestley [1995]. This approach is preferred over the simplified
method suggested by FEMA273 [1997], which does not explicitly account for the
effects of axial pre-compression in the joint. The critical principal tensile and
compressive stresses developed at the beam-column joints were found to be within the
permissible limits in both tension and compression.
5. EXPERIMENTAL PROGRAM ON SOFT-STOREY COLUMNS
In this study, an experimental program on typical soft-storey columns will be
undertaken to develop and verify the analytical column-deformation model. Cantilever
column specimens representative of current constructional practices will be tested to
study the force-displacement behaviour under both monotonic and cyclic loading. The
typical experimental set-up is shown in Figure 5. The newly established Digital Close-
range Photogrammetry Technique [Fraser, 19971 will be used for measuring
deformation and surface strains. One of the key objectives of the experiments is to
evaluate the accuracy of the analytical procedure in resolving deformation into the
respective components including the contributions from flexure, shear and yield
penetration.
Page 17-5
Figure 5. Experimental equipment for cyclic load test
6. CLOSING REMARKS
This paper presents the application of a simple and effective non-linear static procedure
to evaluate the seismic performance of a soft-storey building by comparing the seismic
demand with the capacity. The seismic demand is represented realistically by a recently
developed response spectrum model plotted in ADRS format. The capacity of the soft-
storey is ascertained using a deformation model that has accounted for the effects of
axial compression, flexure, shear, colurnn-end rotation and foundation flexibility. The
uneven sharing of shear forces between the columns and the significant additional shear
demand associated with the strut actions in the masonry infill are amongst the numerous
important issues that have been raised for special attention. An experimental program is
being planned to confirm the accuracy of the analytical procedure.
7. REFERENCES
ACI3 18 (2002) Building Code Requirements for Reinforced Concrete and Commentary,
American Concrete Institute.
Alsiwat, M. and Saatcioglu, M., (1992) Reinforcement Anchorage Slip Under
Monotonic Loading, Journal of Structural Engineering, Vol 118(9), pp 2421-2437.
ASJNZS 1170.4 Draft no.8 (2003) Structure Design Actions - Part 4: Earthquake
Action, sub-committee BD/6/4, January.
AS1 170.4 (1993) Minimum Design Loads on Structures: Part 4: Earthquake Loads.
ATC40 (1996) Seismic Evaluation and Retrofitting of Concrete Buildings, Applied
Technology Council, USA.
FEMA273 (1997) NEHRP Guideline for the Seismic Rehabilitation of Buildings,
Washington DC, Federal Emergency Management Agency, USA.
Page 17-6
Fraser, C., (1997) Some Thoughts on the Emergence of Digital Close-Range
Photogrammetry, President's Medal Address to the Photo'gamrnetric Society, London.
Moehle, J., Elwood K. and Sozen, H., (2002) Gravity Load Collapse of Building Frames
During Earthquakes, Special Publication, Uzumeri Symposium, American Concrete
Institute, Farmington Hills, Michigan.
Panagiotakos, T. and Fardis, M., (2001) Deformations of Reinforced Concrete Members
at Yielding and Ultimate", ACI Structural Journal Vol98(2), pp 135-148.
Park, R. and Paulay, T., (1975) Reinforced Concrete Structures, Wiley.
Priestley, M., (1995) Displacement Based Seismic Assessment of Existing Reinforced
Concrete Buildings, Proceedings of the Pacific Conference on Earthquake Engineering,
V012, pp 225-244.
Priestley, M., Verma, R. and Xiao, Y., (1994) Seismic Shear Strength of Reinforced
Concrete Column, Journal of Structural Engineering ASCE, Vol 120(8), pp 2310-2329.
Vecchio, FJ., and Collin, MP., (1986) The Modified compression Field Theory For
Reinforced Concrete Elements Subjected to Shear, ACI Structural Journal, Vol 83(2),
pp 219-231.
Watson, S., Zahn, F. and Park, R., (1994) Confining Reinforcement for Concrete
Columns, Journal of Structural Engineering, ASCE, Vol 120(6), pp 1798-1824.
Wilson, J. and Lam, N., (2003) A recommended earthquake response spectrum model
for Australia, Australian Journal of Structural Engineering, Institution of Engineers
Australia, Vol5(1), [In press]
Page 17-7
