DISCUSSION FORUM

Effect of weak and soft storeys on seismic performance of

reinforced concrete frames with unreinforced brick infills
Discussion by N. Subramanian

Replies by Patnala V.S. Neelima and R. Pradeep Kumar

This has reference to the paper titled ‘Effect of weak and (GF) soft storey and all the other elements in the higher
soft storeys on seismic performance of reinforced concrete stories are unaffected (GF columns are imposed to large
frames with unreinforced brick infills’, authored by Patnala deformation and also plastic hinges are formed at top and
V.S. Neelima and R. Pradeep Kumar, published in The bottom of the GF columns). In spite of this weakness, this
Indian Concrete Journal (Vol. 90, No. 2, pp. 19-26). kind of arrangement is still preferred in several of our big
cities by architects and developers, since it is convenient
Several past earthquakes have demonstrated that the open to provide covered car park area in the ground floor. Thus
ground storey is vulnerable during earthquakes and is the the basic requirement of safety is compromised in these
cause of total failure of the structure, as shown in Figure 15. structures during earthquake. Hence they should be banned
From this figure, it is clearly seen that the failure is confined at least in Seismic zones VI and V.
to the columns and beam-column joints in the ground floor

46 The Indian Concrete Journal June 2016

DISCUSSION FORUM
Incidentally, after observing the string of recent earthquakes building as feasible, and designed exclusively for 1.5 times the
and volcanic activity in the ‘ring of fire’ region, especially lateral calculated storey shear force”. Thus, considering Clause
at Ecuador and Japan, Los Angeles released publically the 7.10.3(a) of IS 1893(part 1):2002 alone, without considering
addresses of 13,500 soft-storey non-ductile apartments and Clause 7.10.3(b) should be considered as a mistake, and
condos, which need to be retrofitted to resist earthquake surely will not result in a safe structure.
forces (Los Angeles building officials went through
numerous city records and walked block-to-block to identify Reply: The issue raised is true but in practice, there are hardly
these structures. Owners of these building were informed very few constructed buildings in any city which will allow
individually, and a number of them have already begun for the provision of shear walls in open parking area. The
the retrofitting process. These retrofits are explained in analysis is performed keeping in mind the practical issues
FEMA 547,2006, and may cost as much as $130,000).  Similar taking place in site.
initiative was taken by the City of San Francisco in 2013.
The code, however, does not specify the stiffness of columns
Such initiatives, to strengthen existing soft-storey non-
or shear walls, which will compensate for the loss of stiffness
ductile buildings, should be taken in India also, in order to
due to the elimination of infill walls in the ground floor. The
save human lives and eliminate/reduce property damage
discusser is of the opinion that we should provide sufficient
during earthquakes.
stiffness, in order to satisfy the storey drift limitation of
0.004 times the storey height as specified in Clause 7.11.1
In order to avoid failures due to soft-storey effect, an ad hoc
of IS 1893 (part 1):2002. It is not sure whether the increased
Clause 7.10.3 was introduced in IS 1893(part 1):2002, without
column dimensions as per the codal provisions, which were
much research. It is interesting to note that this paper under
provided by the authors in the reply to the discussions and
discussion disputes this clause and concludes that the columns
published in the March 2016 issue of the Journal, satisfied
designed for 2.5 times the storey shears and moments will
the storey drift limitation specified in Clause 7.11.1 of
not perform satisfactorily during earthquakes.
IS 1893(part 1):2002.
The authors seem to have designed only the columns in the
Reply: According to the existing Indian code, Clause 4.20,
soft storey for the increased storey shears and moments,
the definition of soft storey is given as “It is the one in which
whereas Clause 7.10.3(a) of IS 1893(part 1):2002 suggests
the lateral stiffness is less than 70 percent of that in the storey
that columns as well as beams in the soft storey are to be
above or less than 80 percent of the average lateral stiffness
designed for 2.5 times the storey shears and moments.
of the three storeys above”. Hence the frame with and
Replies from the authors without soft storey differ by the criteria of stiffness. In any
recommendation given on soft storey building in code, the
Firstly, the authors are deeply thankful to the reader for going main objective is to match the three parameters of capacity
through the published paper thoroughly and rigorously. of the building; stiffness, strength and ductility. Hence to
The authors appreciate the interest shown by the reader in compensate the loss of stiffness due to elimination of infill
understanding the paper. We put our full effort for responding walls, the open ground storey columns are designed for higher
to the queries. design forces and moments. But the stiffness need not provide
the storey drift limitation of 0.004 times the storey height as
According to the clause 7.10.3(a) of IS 1893:2002, the design
specified in the code. Rather the load carrying capacity and
moments in both the columns and beams are to be increased
seismic behaviour of the improved building should match
by 2.5 times, but increase in design moments of beams
or atleast should be comparable with the building with infill
may lead to strong beam and weak column which is not a
walls.
suggestible solution in earthquake resistant design of RC
buildings. Hence according to “Proposed draft provisions Moreover, in such soft storey GF columns, the confinement
and commentary on Indian Seismic Code IS 1893 (Part 1), reinforcement in the form of shear ties should be provided as
Document No. IITK-GSDMA-EQ05-V4.0”, Clause 7.11.1.2 per IS 13920:1993, at closer intervals throughout the height of
and commentary C7.11.1.2, the columns are only designed the column (and also in the beam-column joint) and should
for increased moments. have 135o hooks (Subramanian, 2011). How this aspect is
considered in the analytical model by the authors? It has to
Also there is a sub-clause (b) which says that ‘besides the
be emphasized that earthquake detailing is very important
columns designed and detailed for the calculated storey shears
and provides ductility to vulnerable columns and beam-
and moments, shear walls should be placed symmetrically on
column joints.
both directions of the building as far away from the centre of the

The Indian Concrete Journal June 2016 47

DISCUSSION FORUM
Reply: The shear reinforcement provided in the columns course, it may not be economical to do a pushover analysis
is designed according to the requirements following the in design offices. It has to be kept in mind that even the
IS 456:2000. The shear stirrups provided in the column pushover analysis is not very accurate and is based on several
are extended through the beam column joint. As the 135o assumptions. Tabeshpour, et al. (2012) provides a discussion
hook is provided for giving sufficient confinement effect to on the different types of modeling and their issues.
the member, this effect is modeled numerically by adding
the additional stiffness to the elements present near the Reply: Non linear static pushover analysis may not be the
stirrups. finest analysis for understanding the seismic behavior of
buildings but it is a better alternative over the linear analysis.
One more problem is that in the current practice, only the Moreover, the present work comments on the design
load due to the infill walls are taken in the analysis and the provisions of buildings subjected to lateral forces for which
frames (either 2D or 3D) are analyzed as bare frames only. Of

48 The Indian Concrete Journal June 2016

DISCUSSION FORUM
the capacity of the building under lateral forces is required mechanism is changed from frame action to predominant
which can be obtained by pushover analysis rather from truss action and the columns will have increased axial forces
other advanced analyses like nonlinear time history analysis, but with reduced bending moments and shear forces (Murty
incremental dynamic analysis etc. and Jain, 2000). Hence, it may be easier to find the behaviour
of these frames, by substituting the infill wall by diagonal
The discusser is not conversant with the Applied Element
struts. Again, several formulae are available for idealizing
Method (AEM) used by the authors. However, it may be
these brick struts. One such idealization is given in Clause
interesting if the authors explain the interface condition,
B-3.3 of the IS 800:2007.
between the infill brickwork and the concrete frame,
considered in the analysis. It is usually better to conduct Reply: As the finite element method is based on the
experimental work to confirm the assumption made in the assumption that the material is continuum, limitations exist in
complex computer analysis. representation of cracks and separation between elements. The
finite element strut modeling assumes the infill wall as a single
Reply: For brick masonry, the material model used was a
strut by which only few parameters responsible for the seismic
composite model that takes into account the brick and mortar
behavior can be assessed. Figure 2 show the comparison
with their respective constitutive relation with elastic and
between the strut modeling and the AEM modeling.
plastic behavior of hardening and softening. Brick springs
were assumed to follow principal stress failure criteria with As discussed by the undersigned in the discussion of an
linear elastic behavior. Once the bricks reach elastic limit, earlier paper published in ICJ on a similar topic, the discusser
normal and shear stresses are not transferred through cracked listed some practical problems while including the effect of
surface in tensile state. Failure modes that come from joint infill brickwork in the analysis (Subramanian, 2013a). A few
participation of unit and mortar in high compressive stress of them are:
is considered by liberalized compression cap. The cohesion
and bond values are constant till the stress first time when Location and size of openings will affect the effect of the
stress exceeds the respective failure envelopes. Figure 1 (a), (b) infill : Most of the internal partition walls in multi-storey
shows the material models used for concrete [10] and steel [10], buildings are half – brick thick (115 mm thickness) walls.
(c), (d), (e) and (f) show the cohesion degradation [1], bond They cannot be assumed to provide any stiffness, as they
degradation [1], failure envelope [1] and softening model [11] will fail immediately during an earthquake. During the 2001
defined at the brick masonry joint. For further details on the Bhuj earthquake (Gujarat, India), even banded half brick
method please refer to the research paper in references [5] in walls collapsed and caused huge loss of lives in residential
the paper under discussion. flats. A reliable method to strengthen these walls, so that
they will perform better during earthquakes, has not been
Several analytical and experimental investigations have been found yet. Although resulting in increased cost, gunited
reported in the literature and show that separation of frame ferrocement, on both sides of the wall, may prevent these
and infill takes place along one diagonal and a compression walls from breaking up. Even in the outer walls, which have
strut forms along the other and the structural load transfer 230 mm thickness, window and door openings, will weaken
the walls, unless RC bands are provided. Even if the opening
are properly modelled and incorporated, it is impossible to
anticipate the changes that may be occur during the life time
of the structure. In fact, the openings in brick walls, if not
properly placed, will even result in short column effect and
affect the performance of RC columns during earthquakes
(Subramanian, 2013b).

Reply: As the present study is carried out in 2d analysis, the

out of plane behavior of the infill wall cannot be simulated.
Hence 3d analysis of the same study has to be carried. As
suggested by the discusser, the effect of opening on seismic
behavior of RC frames with masonry infill walls will be dealt
in the future work.

Dynamic load and inelastic behaviour: It is well known

that behaviour of the infill is brittle in nature and once it

The Indian Concrete Journal June 2016 49

DISCUSSION FORUM
cracks, the frame stiffness will drop drastically. Also, during Reply: As suggested by the discusser, the effect of infill
the first few cycles of earthquake, the infill brick wall will material property variation on seismic behavior of RC frames
collapse completely (in fact many injuries are attributed to with masonry infill walls will be dealt in the future work.
the flying debris of brick walls).
Quality of material and workmanship: Some of the blocks
Reply: Same as the previous reply to above query. such as AAC and CLC are manufactured in the factory and
have better quality control than others like clay bricks and
Infills may result in non-ductile performance: The increased flyash bricks. Very poor quality bricks are found to be the
stiffness of the building due to the presence of infills may cause of the failure of infilled walls on several occasions.
reduce the ability of the frame to flex and deform. In ductile Similarly, workmanship may also vary considerably,
RC frames, masonry infills may prevent the primary frame resulting in weak joints, susceptible for cracks and ultimate
elements (i.e., columns and beams) from responding in a collapse.
ductile manner - instead, such structures may show a non-
ductile (brittle) performance. This may culminate with a Reply: As workmanship cannot be modeled, this is the
sudden and dramatic failure (Murty et al, 2006) limitation of the numerical modeling

Reply: Though the discussers view is valid, infill walls are Richer mortar has to be used: Richer cement-sand mortar
the better elements for dissipation of energy input to the of 1: 4 mixture (1 part cement by 4 parts of sand) makes the
buildings. Hence the seismic performance of brick infill walls masonry stronger against earthquake shaking as compared
have to clearly understood in relation with the RC frames. with the usual 1:6 mortar used in such construction, by a
factor of 2.5 to 3.0
Infill material property variation: There are many infill
material available these days - clay bricks, flyash bricks, solid Reply: As suggested by the discusser, the effect of mortar
concrete blocks, hollow concrete blocks, Aerated Concrete richness on seismic behavior of RC frames with masonry infill
(AAC) Blocks, Cellular Lightweight Concrete (CLC) blocks, walls will be dealt in the future work.
Perforated Clay Blocks, Compressed Stabilized Earth blocks,
Some more additional queries on the paper are:
etc. Each has different material properties. Considerable
research is required to identify their behaviour under cyclic
Their analysis shows that there is huge increase in load
loading from earthquake and also the effects of openings in
carrying capacity of the structure, if the structure is having
such walls.

50 The Indian Concrete Journal June 2016

DISCUSSION FORUM

weak storey in the first floor, when compared to the ground which is different from concentrated plasticity model used
floor. However, in actual earthquakes, weak stories present in FEM.
in the structure, irrespective of the level, is considered to
Which formula was used to calculate the fundamental
cause pan caking failure (see Figure 16).
natural period of the structure?
Reply: Definitely, the load carrying capacity of the building
Reply: According to IS 1893:2002, the formula provided for
with weak storey at first floor is more when compared to the
RC buildings with infill walls is used.
building with weak storey at the ground floor. The comparison
is relative. Moreover, the presence of infill gives a higher load In Table 2, they have shown that the maximum base shear
carrying capacity only in the initial stages of loading but very for the structure with soft storey has a maximum base shear
soon the load carrying capacity decreases suddenly. This can of only 160 kN, as compared to 4673 kN for the structure
be seen in any of the pushover curves for buildings with infill without soft storey. Similarly, the modified structure
walls. The interpretation drawn from the analyses carried out considering 2.5 times the shear and moments, as per Clause
is that with the change in weak storey from ground floor to 7.10.3(a) of IS 1893(part 1):2002, has a maximum base shear
the top floor, the likelihood of the collapse of the structure or of only 578 kN. Design seismic base shear is dependent on
the concentrated failure of one floor (i.e., pan cake collapse) Ah (design horizontal acceleration spectrum value) and W
can be decreased. (seismic weight of the building) only. Will the removal of
infill wall only in the GF, result in so much reduction in the
In their analysis have they considered reduced stiffness
design seismic base shear?
due to cracked column and beam section properties, which
will affect the results considerably (see Section 10.10.4 of
Reply: According to the code, the formula for base shear
ACI 318-11)?
represents the design base shear but not the maximum
base shear. Moreover, the design base shear given in code
Reply: Yes. As the Applied Element Method (AEM) is
is obtained from linear analysis. The nonlinearity of the
continuum modeling, the reduction in stiffness due to failure
building is represented by a factor ‘R’ response reduction
of each and every finite element is considered in the analysis
factor. Whereas the maximum base shear of the building is
and the new stiffness is used in every step for finding the
obtained from non linear static pushover analysis. This can
load carrying capacity. AEM uses distributed plasticity model
be clearly illustrated in Figure 3.

The Indian Concrete Journal June 2016 51

DISCUSSION FORUM
References 14. Subramanian, N., Design of Reinforced Concrete Structures, Oxford
University Press, 2013b, 872 pp.
8. Neelima, P. V. S., and Pradeep Kumar, R. “Effect of weak and soft 15. Murty, C.V.R., Brzev, S., Faison, H., Comartin, C.D. and Irfanoglu,
storeys on seismic performance of RC frames with unreinforced A., “AT RISK: The Seismic Performance of Reinforced Concrete
brick infills, The Indian Concrete Journal, Vol. 90, No.2, Feb. 2016, Frame Buildings with Masonry Infill Walls”, A Tutorial Developed
pp. 19-26. by a committee of the World Housing Encyclopedia, Earthquake
9. Subramanian, R., Discussion on “Effect of weak and soft storeys on
Engineering Research Institute, Oakland, California, November
seismic performance of RC frames with unreinforced brick infills, The
2006, 83 pp.
Indian Concrete Journal, Vol. 90, No.3, Mar. 2016, pp. 42-43.
10. FEMA 547, Techniques for the Seismic Rehabilitation of Existing
Buildings, Federal Emergency Management Agency, Washington
D.C., 2006, 571 pp Discussion by
11. Tabeshpour, M.R., Azad, A., and Golafshani, A.A., “Seismic Behavior
and Retrofit of Infilled Frames”, in Earthquake-Resistant Structures Dr. N. Subramanian, 23, Napa Valley Road, Gaithersburg, MD
- Design, Assessment and Rehabilitation, Moustafa A., Ed., InTech,
20878, USA
Croatia, ISBN: 978-953-51-0123-9, 2012, pp.279-306 [http://www.
intechopen.com/books/earthquake-resistant-structures-design-
assessment-and-rehabilitation] Replies by
12. Subramanian, N., “Design of confinement reinforcement for RC
columns”, The Indian Concrete Journal, Aug. 2011, Vol. 85, No. 8, Patnala V. S. Neelima and Professor R. Pradeep Kumar,
pp. 25-36.
Earthquake Engineering Research Centre (EERC), International
13. Subramanian, N., Discussion on “Effect of brick masonry infill in
seismic evaluation of an existing RC building” by Vishal P. Jamnekar Institute of Information Technology, Hyderabad (IIITH),
and D.J. Chaudhari, The Indian Concrete Journal, Dec. 2013a, Vol. Gachibowli, Hyderabad, Telangana 500032, India.
87, No. 12, pp. 42-45.

A CD on Repair and Rehabilitation

An Icj Compilation
• Everything you want to know in repair and
rehabilitation

• 45 selected papers published in ICJ during the

recent past

• Indian case studies from those who were associated

• Contributions from world-renowned experts like

P. Emmons, D. Kaminetsky, Dr. G. M. Sabnis,
A. Vaysburd, N.P. Mailvaganam, etc.
browse pages in
Price: 600 only Flipbook format

To purchase any CD, please pay online at www.icjonline.com or send a cheque / DD of Rs. 600 in favour of 'ACC Limited' to:
The Publication Manager, The Indian Concrete Journal, ACC Limited, L.B.S. Marg, Next to Eternity Mall,
Near Teen Haath Naka, Thane (W) 400 604. Tel: +91(22) 3302 7646 E-mail: [email protected] Website: www.icjonline.com

Online payment facility available at www.icjonline.com

Cost of any CD is Rs. 600 inclusive of postage and handling. All Cheques / Demand Drafts should be drawn in favour of 'ACC Limited' and
payable at Mumbai. For non-mumbai-clearing cheques, please add Rs. 100 (as bank clearing charges).

52 The Indian Concrete Journal June 2016