th

The 14 World Conference on Earthquake Engineering

October 12-17, 2008, Beijing, China

BASE SHEAR REDISTRIBUTION BETWEEN THE R/C DUAL SYSTEM

STRUCTURAL COMPONENTS
1 2 3
V. Sigmund and I.Guljas and M.Hadzima-Nyarko
1 2 3
Professor, Assoc. professor Assistant
University of Osijek, Faculty of Civil Engineering,Drinska 16a,31000 Osijek, Croatia
Email:sigmund@gfos.hr

ABSTRACT :

The experience provided by the recent earthquakes showed that buildings with structural walls have had
satisfactory performance for collapse prevention. Conversely, some of the dual system buildings (wall frame)
were rated as severely damaged therefore implying an urgent need for strength assignments redefinition
between the structural components. The elastic response of dual buildings is mainly governed by the wall
response and the frame contribution could be neglected (bellow 15% of the base shear what makes them a
secondary member). In the inelastic range the situation could drastically change and frame contribution
increases (even up to the 50%) but these elements are then under-designed and not capable to carry the
increased load.

A set of dual building structures were selected, designed and analyzed. The model buildings were 4, 7 and 10
stories high with two frames and one wall-frame in longitudinal and four frames in the transverse direction.
Column dimension varied thus causing stiffness and strength redistribution among the wall and frame, in elastic
and inelastic range. Seismic analysis including elastic, nonlinear static (push-over) and nonlinear dynamic
analysis have been done. Observed were base shear distribution, displacement and ductility demands of the wall
and frames through system degradation. The results have shown that wall contribution diminishes as
deformations are increasing and its share could go up to almost 50% what required high ductility of the frame
elements. That they do not have if they were originally designed as “secondary” seismic elements according to
the material Eurocodes.

r/c dual system, Eurocodes, earthquake, primary and secondary seismic

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

1. MODEL BUILDINGS

Modern seismic codes (EN 1998-1) make a distinction between the primary and secondary seismic elements, as far
as their role and contribution to earthquake resistance of the building is concerned. “Primary seismic members”
should be modeled in the structural analysis, designed and detailed for earthquake resistance. Secondary seismic
elements only need to satisfy the rules of the material codes plus the requirement that they maintain support of
gravity loads when subjected to most adverse displacements and deformations induced in them in the seismic design
situation.

In the dual wall-frame buildings, wall easily, due to its

higher stiffness in the elastic analysis, attracts higher
portion of the horizontal loads. That makes it a “primary
seismic element” and frames are designed for practically
no horizontal loads. When the frames are exposed to
expected non-linear displacements of the system then
induced forces are much greater than the ones obtained
during the linear elastic analysis. If the frames were
designed for these forces than they are no longer
“secondary seismic member”. So, the distinction between
primary and secondary elements is lost as well as the idea
to simplify the design by not considering frames in the
structural model for seismic analysis of the building.

Stories: 4
Wall/frame
clmn beam beff (cm) Wall length Wall -thick
(cm) (cm) (cm) (cm)
40x40 30/50 111,2 540 20
50x50 30/50 110,5 550 20
60x60 30/50 109,8 560 20
Stories: 7
Wall/frame
clmn beam beff (cm) Wall length Wall -thick
(cm) (cm) (cm) (cm)
50x50 30/50 110,5 550 20
60x60 30/50 109,8 560 20
70x70 30/50 109,1 570 20
Stories: 10
Wall/frame
clmn beam beff (cm) Wall length Wall -thick
(cm) (cm) (cm) (cm)
60x60 30/50 109,8 560 20
70x70 30/50 109,1 570 20
80x80 30/50 108,4 580 20

Table 1 Geometrical data of the models

Figure 1 4-story model building

In order to point this problem out, a set of model buildings has been defined, designed and verified. The model
buildings represent the well-known “Tsukuba building” which in the lateral direction has dual structural system
consisting of boundary frames integrally linked to the central structural wall (5.00*0.20m) by moment-resisting
beams. In transverse direction the building is a pure frame building having three moment-resisting frames (Fig.
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

1). The model buildings have 4, 7 and 10 stories with standard loading for business/dwelling buildings (g=9
kN/m2 and q=3.5kN/m2) located in the IX seismic zone (with design PGA=0,3g) grounded on the ground type
„B“. Frame-beam dimensions were kept constant while frame-column dimensions varied as follows: a) 40x40;
50x50 and 60x60cm; b) 50x50; 60x60 and 70x70cm; c) 60x60; 70x70 and 80x80cm for 4, 7 and 10 stories
respectively. Wall/floor area is kept constant at 0.36% (wall is 5,00x0,20m). All buildings were designed by
linear-elastic modal analysis methods according to the Eurocodes EN 2 and EN 8 and for high ductility (DC H)
of their elements.

2. NON-LINEAR NUMERICAL ANALYSIS

Numerical model for calculating the nonlinear response of reinforced/concrete building structures was
developed and is explained in detail in Lopez (1988). Structure is defined in terms of its geometry and
moment-curvature relationship for its individual elements. The frame elements consist of the linearly
elastic member with nonlinearities concentrated in two nonlinear springs at the member ends (Giberson,
1969). Nonlinear response of wall elements includes flexure and shear components. Each wall member
consists of several subelements so that each subelement can be subjected to a different stage of inelastic
action in order to allow inelastic propagation through the story height. Force-deformation relationship of
the members for monotonically increasing loads have been defined using the empirical approach. This
combination has been verified as the one giving very good results in Sigmund (2003). Shear force and
shear displacement relationship were calculated by taking only reinforcement for shear carrying
capacity of the section.

2.1. Non-linear static

10000 (push-over) analysis
For evaluation of the attained
4_40

9000 4_50
4_60 structural capacity static
8000 7_50
7_60
non-linear pushover analysis
7000 7_70 has been performed.
10_60
Incremental horizontal loading
base shear (kN)

6000
10_70

5000
10_80 was applied in a form that
represents the first structural
4000
mode (triangular lateral load).
3000 Calculated relationship between
2000 the base shear and roof level
1000
displacement represents the
overall nonlinear characteristics
0
0 2 4 6 8 10 12 14 16 18 20
of the structure and its
roof displacement (cm) structural capacity.

Figure 2 Base-shear Roof-displacement of the buildings

The initial elastic stiffness of the structures did not differ significantly by increasing the frame-column
dimensions which indicates that wall stiffness plays a major role in the elastic phase. Yield and
post-yield stiffness increased from 10 to 15% . In the Figure 2 and Table 2 presented are the results of the
push-over analysis for the 4, 7 and 10 stories building. Ultimate base shear is determined by the roof displacement
equal to 2% of the building height. Base-shear at the crack opening is set as the 25% of the ultimate base shear.
Generally, increase of the column cross-section (frame stiffness and strength) increased the overall structural
lateral stability.

Wall-response dominated the overall structural behavior in all model buildings. Higher buildings (with 7 and 10
stories) and stronger columns have shown unfavorable local behavior in the wall and frame-columns.
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

Development of the plastic hinges and required local ductility is shown on the Figures 3, 4 and 5 for 4, 7 and 10
story buildings respectively. Plastic hinges were opening at the wall bottom (when the frame contribution is
neglectable), but also on the first and second floor (for frames having stiffer columns) which could be very
unfavorable. That required higher column ductility, which is not provided by default in a „secondary
seismic member“.
Table 2 Base-shear and Roof-level displacement characteristic values
Stories 4 7 10
Base shear Column section Column section Column section
(kN);
Roof
40x40 50x50 60x60 50x50 60x60 70x70 60x60 70x70 80x80
displacement
(cm)
BSC (kN) 1321 1501 1681 1752 1902 2002 2402 2752 3251
BSY(kN) 4570 5090 5450 5480 5700 5900 6200 6950 7555
BSU (kN) 5254 5945 6695 7058 7559 7959 9607 11008 13009
dc (cm) 0,46 0,45 0,45 1,34 1,32 1,32 2,58 2,72 2,96
dy (cm) 6,22 5,95 4,80 11,3 10,6 10,4 15,7 14,8 14,1
du (cm) 25,59 25,86 25,67 43,17 43,12 43,55 61,34 61,63 61,50

18 27 5 12 9 11 18 27
34 14 14 24 33 12 34 14 14 24
24 24 24 24
17 30 23 29 25 22 3 13 7 7 17 30 23 29 25 22
32 33 32 33

16 26 20 25 22 21 2 13 8 6 16 26 20 25 22 21

15 26 19 25 22 19 4 14 10 6 15 26 19 25 22 19

29 31 31 28 29 1 29 29 31 31 28

Figure 3 Opening of the plastic joints and required ductility, 4 stories-50x50cm columns
20 36 31 37 26 29 13 14 9 19 20 36 31 37 26 29
26 26 34 41 35 26 26 34

23 24 14 25 13 28 7 13 6 16 23 24 14 25 13 28

21 21 8 22 8 27 5 12 4 14 21 21 9 22 8 27

20 18 4 19 5 26 4 11 3 12 20 18 4 20 7 27

17 22 9 21 11 24 4 14 8 11 17 22 9 18 7 22

43 10 2
19 23 12 23 13 26 8 16 11 14 19 23 27 39 32 30
16 7
33 44 42
25 28 19 28 20 30 15 22 18 21 25 28 18 25 15 28
45

37 40 40 37 38 1 38 37 40 41 39

B E A M P L A S T IC J O IN T
C O L U M N P L A S T IC J O IN T
D U C T .= 1 2 3 4 5 6 7 8 9 W A L L P L A S T IC J O IN T

Figure 4 Opening of the plastic joints and required ductility, 7 stories-60x60cm columns
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

2 16 38 38 47 29 17 9 37 33 39 32 40 37 47

43 56 60 58 56
1
33 52 13 29 10 46 29 17 7 33 38 27 14 29 10 46
3
54
33 25 12 28 9 45 27 14 5 32 37 25 11 28 9 45

35 23 8 24 5 43 24 11 4 30 34 23 8 24 5 43

25 26 15 25 16 37 19 14 7 26 26 26 15 25 16 37

24 25 13 24 15 36 18 13 6 26 24 25 13 24 15 36

25 25 14 24 16 37 19 14 7 26 25 25 14 24 16 37

27 32 17 30 20 36 19 22 13 25 27 32 17 30 20 36

53
32 36 23 35 25 41 25 28 21 31 32 36 23 35 25 41
55
42
45 49 35 48 37 51 38 40 33 44 45 49 35 48 37 51
50

58 61 61 57 59 13 59 58 61 61 57

Figure 5 Opening of the plastic joints and required ductility, 7 stories-60x60cm columns

2.1.1 Distribution of the base shear among the wall and columns
Distribution of the base shear among the wall and frame-columns is visible on the Figures 6, 7 and 8. It is obvious
that the wall contribution diminishes as the plastic deformation increases. In the elastic range wall takes from 83
to 93% of the toal base shear. At the ultimate state it takes only 48 to 63% of the base shear which means that
frame elements do not represent a „secondary seismic member“ as inelastic deformation increases.
10 10 10
0 0 0
9 FRAM 9 9
FRAME
0
E 0 FRAM 0
8 8 8
0
E 0
0
7 7 7
0 0 0
6 6 6
0 WALL 0
%
5 % 0 %
5
5
0 0 WALL
0 WALL
4 4 4
0 0 0
3 3 3
0 0 0
2 2 2
0 0 0
1 1 1
0 0 0
BS BS
0 BS BS BS 0 BS BS
0 BS y u
0 100 200 300 400 500 BS 0 100 200 300 400 500 600
c y u 0 100 c
200 300 400 500 y 600 u 700 c
0 0 Base 0Shear 0 0 0 0 0
Base 0
Shear 0 0 0 0 0 Base0Shear [kN]0 0 0
[kN] [kN]

a) b) c)
Figure 6 Distribution of the base shear among the wall and columns for 4 story building and column dimensions
of a) 40x40 cm; b) 50x50 cm; c) 60x60 cm

100 100 100

90 FRAME 90 FRAME 90
FRAME
80 80 80

70 70 70

60 60 60
% % %
50 50 50
WALL WALL WALL
40 40 40

30 30 30

20 20 20

10 10 10

BSc BSy BSu BSc BSc BSy BSu

0 0 0
0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 BSy 6000 7000 BSu 0 1000 2000 3000 4000 5000 6000 7000 8000
Base Shear [kN] Base Shear [kN] Base Shear [kN]

a) b) c)
Figure 7 Distribution of the base shear among the wall and columns for 7 story building and column dimensions
of a) 50x50 cm; b) 60x60 cm; c) 70x70 cm
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

100 100 100

90 90 90
FRAME FRAME FRAME
80 80 80

70 70 70

60 60 60
% % %
50 50 50
WALL WALL WALL
40 40 40

30 30 30

20 20 20

10 10 10

0 BSu 0 0
0 2000BSc 4000 6000 BSy 8000 10000 0 2000 BSc 4000 6000 BSy 8000 10000 BSu 0 2000 BSc 4000 6000 BSy8000 10000 12000 BSu

Base Shear [kN] Base Shear [kN] Base Shear [kN]

a) b) c)
Figure 8 Distribution of the base shear among the wall and columns for 10 story building and column dimensions
of a) 60x60 cm; b) 70x70 cm; c) 80x80 cm

For 4-story model frame contribution in the elastic range was neglectable (from 8 to 19%). By approaching the
inelastic deformation of 1.5% of the MDR its contribution increased to 33 to 52%.
For 7-story model frame contribution in the elastic range was from 14 to 22%. By approaching the inelastic
deformation of 1.5% of the MDR its contribution increased to 26 to 34%.
For 10-story model frame contribution in the elastic range was from 21 to 31%). By approaching the inelastic
deformation of 1.5% of the MDR its contribution remained about the same.
Observation of the capacity curves indicated that discrepancy between the elastic and inelastic frame contribution
is bigger when its contribution, according to the elastic analysis, could be neglected.

2.2. Non-linear dynamic analysis

Nonlinear dynamic time history analysis of the models was calculated using LARZWD (1992) and a set of three
recorded ground motions that covered the frequency range of interest: Bar N-S and Petrovac N-S recorded during
the 1979 Monte-Negro earthquake and El Centro N-S recorded at 1940. Maximum accelerations were scaled so
that ground motion spectral intensities were similar for the same earthquake zone (IX zone MCS). The set of three
different ground motions was used.

MDR (%) for 4 story model IDR for 4 story model

4
4

3
3

2 2

1 1
IDR
MDR (% )
0 0
0 0,5 1 1,5 2 0 0,5 1 1,5 2
40x40-Petrovac 50x50-Petrovac 60x60-Petrovac 40x40-Petrovac 50x50-Petrovac 60x60-Petrovac
40x40-Bar 50x50-Bar 60x60-Bar 40x40-Bar 50x50-Bar 60x60-Bar

Figure 9 Calculated MDR and IDR of the 4-story model for Petrovac & Bar records
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

MDR (%) for 7 story model IDR for 7 story model

7 7
6 6
5 5
4 4
3 3
2 2
MDR (%) IDR
1 1
0 0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6
50x50-Petrovac 60x60-Petrovac 70x70-Petrovac 50x50-Petrovac 60x60-Petrovac 70x70-Petrovac
50x50-Bar 60x60-Bar 70x70-Bar 50x50-Bar 60x60-Bar 70x70-Bar

Figure 10 Calculated MDR and IDR of the 7-story model for Petrovac & Bar records

MDR (%) for 10 story model IDR for 10 story model

10 10
9 9
8 8
7 7
6 6
5 5
4 4
3 3
MDR (%) IDR
2 2
1 1
0 0
0 0,2 0,4 0,6 0,8 1 1,2 0 0,2 0,4 0,6 0,8 1 1,2 1,4
60x60-Petrovac 70x70-Petrovac 80x80-Petrovac 60x60-Petrovac 70x70-Petrovac 80x80-Petrovac
60x60-Bar 70x70-Bar 80x80-Bar 60x60-Bar 70x70-Bar 80x80-Bar

Figure 11 Calculated MDR and IDR of the 10-story model for Petrovac & Bar records

90
80
70
60
50
40
30
20
10
0
60x60 70x70 80x80
wall10 77 76 74
frame10 23 24 26
wall7 79 76 69
frame7 21 24 31
wall4 80 69 58
frame4 20 31 42

Figure 12 Calculated distribution of the base-shear forces between the wal and frame-elements

During the inelastic dynamic analysis distribution of the base shear among the wall and frames depended on the
frequency characteristics of the excitation that exciteed all modes. Frames took over from 20 to 42% for 4 story
building, from 21 to 31% for 7 story building and from 23 to 26% for 10 story building in the average for all
 th
The 14 World Conference on Earthquake Engineering
October 12-17, 2008, Beijing, China

three excitations. Frame-element contribution has been higher in the structures that had undergone higher
inelastic deformation (4-story building).

3. SUMMARY

In order to point out the problem in distinguishing between the „primary“ and „secondary“ seismic members in
dula wall-frame buildings 9 model structures having 4, 7 and 10 stories have been designed according to the
Eurocodes EN 1992 and 1998 by linear-elastic analysis and designed according to the EN 1998 capacity design
rules. Wall/floor area was kept constant at 0,36%. According to the linear elastic analysis, the wall could be
designed as „primary seismic element“ according to the EC8 rules, as it takes over from 85 to 94% of the total
base shear for 4 and 7 story buildings. Therefore, the frame-columns could be designed as „secondary“ seismic
elements for vertical loadings only and according to the EC2.
Non-linear static (push-over) and dynamic (time hostory) analysis of the model structures were done. As the
structures were enetering into nonlinear range, frames were taking over bigger part of the base shear and the wall
took over from 48 to 83% of the total base shear. Contribution of the higher modes, that was obvious in the
non-linear dynamic response, contributed to the lower influence of the wall. Due to the frame-wall interaction,
frame sections have had high ductility demand although they had been designed for gravity loadings only.
Therefore, it seems that in dual (frame-wall) buildingsa, there is no clear distincion among the „primary“ and
„secondary“ seismic elements and that neglecting of the frame-elements contribution could endanger the overall
structural stability. Although wall dominated the dynamic response, contribution of the higher modes started a
pronounced interaction among the wall and frame-elements causing higher than expected non-linear behavior of
the both.

REFERENCES

Sigmund, V., Bošnjak-Klečina, M., Guljaš, I., Stanić, A.. (2000), A comparisson between the buildings designed
according to the HRN and EC8, Građevinar, 52:7, 379-389.
Sigmund, V., Guljaš, I., Stanić, A.(2002)., Non-linear response of structures designed according to the HRN and
EC8, Građevinar, 54:1, 1-14;
EC8: (2004), Design of structures for earthquake resistance, CEN, Brussel, 2004;
Lopez, R., Sozen, M. A. (1988, 1992), LARZWD and LARZWS Computer Programs for Nonlinear Analysis of
Planar Reinforced Concrete Structures Incorporating Frame and walls, University of Illinois, Urbana
Sigmund,V, Brana, P, Herman, K. (2000), Two methods for estimating nonlinear characteristics of r/c frame
members, The Third Japan-Turkey Workshop on Earthquake Engineering, Vol. 1, 87-98., Istanbul, Turkey
Sigmund V., Herman, K., Guljaš, I.(2000), An evaluation of the displacement controlled design procedures, 12
World Conference on Earthquake Engineering, Auckland, New Zealand
Aoyama, H. (1981), A method for the evaluation of the seismic capacity of existing r/c buildings in Japan,
NZNSEE, 14:3
Y.Inada (1993), Force displacement models for solid shear walls of reactor buildings, Earthquake resistance of
reinforced concrete structures, University of Tokyo, Japan
Stanić, A., Sigmund, V., Guljaš, I.(2003), Behavior of the walls under in-plane horizontal loadings, fib-2003
Symposium, Athens, Greece
Stanić, A.; Sigmund, V.; Guljaš, I.(2004), Seismic capacity of structural members using neural networks, XIII
WCEE, Vancouver, Canada