You are on page 1of 14

13th World Conference on Earthquake Engineering

Vancouver, B.C., Canada


August 1-6, 2004
Paper No. 53

RESEARCH ON ADVANCED STRUCTURAL FRAMING SYSTEM FOR


HIGH-RISE OFFICE BUILDINGS

Naoki NIWA1, Seiji AKIYAMA2, Satoru NAGAI3,


Kazunari MAKIBE4 and Akio TOMITA5

SUMMARY

We present an advanced structural framing system, which can construct cost-efficient high-rise office
buildings with high additional value. Main components are comprised of, (1) earthquake-resisting core
walls with boundary beams, which can bear almost all of the earthquake forces, (2) outer frames and (3)
the inversed haunch beams of office areas released from earthquake force. These characteristics give
flexibility to building planning and future possible renovations. The seismic response analysis results
illustrated that the earthquake-resistance standards of Japan, as a severely seismic country, could be
satisfied, and the boundary beams reduced the seismic response. The loading tests confirmed that the
shear strength and bending characteristics of the earthquake-resisting walls with built-in steel could be
evaluated by conventional design equations for reinforced concrete earthquake-resisting walls. It was also
verified that the boundary beams proposed had a large equivalent damping factor and could decrease
damage compared with boundary beams of normal cross-sections.

INTRODUCTION

Sustainability of current and future office buildings has become an important issue in terms of improving
asset value and contributing to future society. Some methods of accomplishing sustainability are
separation of the skeleton and infill, and integration of earthquake-resisting elements into building
structures. Earthquake-resisting element integration, for example earthquake-resisting walls in the core
and also tubular outer structure, have already been done. However, these methods alone are insufficient to
ensure safety against large earthquakes in severe seismic countries like Japan. Furthermore, it is
expensive to separate earthquake-resisting members and permanent load support members.
The authors propose an advanced structural framing system for constructing highly value-added
high-rise office buildings at low cost. Main components of the proposed structural framing system are
comprised of, earthquake-resisting core walls, which can bear almost all of the earthquake forces, and
outer frames. The core walls are connected to each other by boundary beams working as dampers. These

1
Supervisory Research Engineer, Dr. Eng., Kajima Corp., Tokyo, JP Email: niwan@kajima.com
2
Supervisory Research Engineer, Kajima Corporation, Tokyo, JP
3
Senior Research Engineer, Kajima Corporation, Tokyo, JP
4
Chief Engineer, Kajima Corporation, Tokyo, JP
5
Group Leader, Kajima Corporation, Tokyo, JP
structural components release the floor framing of office areas from earthquake force; thus, it can be
supported by simpler structural components, e.g. inversed haunch beams. These characteristics give
flexibility to building planning and future possible renovations. Exactly, its system creates a tower
building with multi-function by hybrid structure. (hereinafter called HMT for short) It is noted that HMT
can be applied to hybrid buildings such as those where the upper stories are a hotel, etc.
In this paper, we report the composition of the HMT and, seismic response analysis results for a model
building. Then, the loading test results of the earthquake resisting core walls with built-in steel and the
boundary beams for the purpose of realizing the HMT are illustrated.

COMPOSITION AND MERIT OF HMT

The HMTs composition is shown in Figure 1. The outer frames are composed of the efficient structural
combination, i.e. reinforced concrete columns and steel beams. The core wall system is composed of
reinforced concrete earthquake-resisting walls and boundary beams that tie their walls together. In the
earthquake-resisting wall, steels are built in as bending resistance reinforcement. The boundary beams are
given the engineering cementitious composites with polyvinyl alcohol (hereinafter called PVA-ECC[1]),
X-shaped steel bars, to achieve high energy dissipation and to minimize damage for the beams function.
The floor framing of the office areas between the outer frames and the core walls can be designed only for
permanent load. Therefore, it can be supported by simpler structural components, e.g. inversed haunch
beams, which the end of the steel beams were slenderized.
4.2
Boundary beams 25FL

Earthquake resisting core walls


with build-in steel

20FL
6.4

B2
15FL
W2
107.6m

Boundary beams
51.2m

Inversed haunch beams


10FL
6.4

Y 5FL
6.4m 6.4
5.5 5.5 4.2

51.2m
X
Outer frames
:Columns is reinforced concrete, 1FL
beams is steel
19.2m 6.4 6.4 19.2
51.2m
Figure 1. HMTs composition and subject building
Accordingly, HMT creates the merits of both planning and facility equipment as follows:
(1) Three-dimensional free space: HMT realizes office space that ensures plane and vertical variability.
Because changes can be made to meet future business requirements, the asset value of the building is
improved. For example, it becomes feasible to maintain tenant-dedicated elevators and stairs, halls
using the two-story space and free layout of well space (see Figure 2).
(2) Free module: The efficient core planning can be created, since the core module is separated from the
main module. Therefore the ratio of the office area to the total area can be augmented. For example, a
compact planning can be efficiently made by setting the main module span to 7.2m and the core
module span to 6.4m to match the elevator module.
(3) Free mechanical space: The inversed haunch beams can create flexibility of ducting works. Therefore
the equipment space of the air conditional systems is reduced by 25% compared to the concrete filled
steel tube structure. For example, in the conventional structure, if a duct is threaded through a beam,
it has to be divided due to the constraint of the opening area. If HMT is employed, the ducting can
freely pass under the beam, thus enabling efficient duct laying while maintaining a larger area (see
Figure 3).

Tenant -
dedicated Tenant-dedicated
elevator stairs

Freedom of space

Well space
Changability Changability
of floor height of floor area
Conferention
room
Freedom of use

Show auditorium exhibition


room
Hall

Figure 2. Three-dimensional free space

Inversed haunch beams Binding beam Slab


Floor level

Return duct Ceiling Haunch Supply duct


Figure 3. Free mechanical space
KEY TECHNOLOGY OF HMT

The characteristics of individual key techniques for the HMT are described below.

Outer frames
The composition of the outer frames is shown in Figure 4. They comprised efficiently hybrid structure of
reinforced concrete columns and steel beams. The reinforced concrete provided strength against vertical
force applied to the column, and the steel provided strength against bending force applied to the beam.
Because the outer frames bear a little of the earthquake force, they can be generally designed for
permanent load. Accordingly, the beams can be used the same member in all the layers. Precast columns
with the same bar arrangement could be used, in which the compression strength of concrete was
increased for the lower stories. Steel plates were installed in the column at the beam flange levels. These
panels and steel beams were jointed with high tension bolts. Therefore, this structure streamlined
construction and saved cost.

Reinfored concrete column

Steel beam

High tension bolt

Steel plate of panel


(a) Cross-section of panel (b) Vertical view of panel
Figure 4. Composition of the outer frame

Earthquake-resisting walls with built-in steel


Cross-sections of conventional reinforced concrete earthquake-resisting wall and earthquake-resisting
wall with built-in steel are shown in Figure 5. Many large-diameter bending resistance reinforcing bars
are arranged at each end of the wall, if there are designed by a conventional reinforced-concrete wall.
Vertically connecting mechanical joints increase cost and construction time.

50-D41 Isc:Interval of stress center

Fc60
BH-4009504545
70-D41 FB2PLs-1950200
Isc:4.75

Isc:5.06
(1.00)

(1.07)
0.7m

6.0m

6.0m
0.7m

12-D41 12-D25
Wall reinforcement Wall reinforcement
lateral D19@100 lateral D19@100
vertical D16@200 vertical D16@200
BH-4006504545
50-D41 FB 2PLs-1950200
0.7m 0.7m
5.0m 5.0m
(a)Cross-sections of conventional reinforced (b)Cross-section of earthquake-resisting wall
concrete earthquake-resisting wall with built-in steel
Figure 5. Conventional reinforced-concrete wall and earthquake-resisting wall with built-in steel
Furthermore, because a large number of the reinforcing bars is required, bending performance is impaired
where these reinforcing bars to resist bending are located near the center of the cross-section. To solve
these problems, these reinforcing bars were replaced with steel frames, bending resistance members were
integrated and joints that used reinforcing bars were removed. Moreover, the bar arrangement for the wall
was made with lap joints with small-diameter reinforcing bars under 19mm. Therefore, this improved the
bending performance and streamlined construction.

Boundary beams
The composition of the boundary beam is shown in Figure 6. The energy dissipation capability was
improved by converting second-step reinforcing bars to X-shaped steel bars. The PVA-ECC is mixed
vinylon fibers (length 12mm and diameter 0.04mm) into the cement mortar. This PVA-ECC is capable of
maintaining strength up to a tensile strain of 3%, and is expected to ensure fine cracks and decrease
damage. PVA-ECC is a kind of high performance fiber reinforced cementitious composite (HPFRCC).
This boundary beams have the function of a seismic response control device. Energy dissipation is
realized from the yield of the reinforcing bars caused by deformation of the core walls during an
earthquake.

PVA-ECC PVA-ECC X-shaped steel bar

X-shaped Core wall Core wall


steel bar

(a) Cross-section (b) Vertical view


Figure 6. Composition of the boundary beam

SIMULATION ANALYSIS FOR MODEL BUILDING

Analysis condition
To clarify the feasibility of the HMT, its safety against earthquakes was verified using a simulation
analysis according to the seismic design standard of Japan. The subject building was a 25-story office
building about 100m heights, as shown in Figures 1. A representative cross-section of the core wall of the
lowest story, a cross-section of the boundary beam of each story and a cross-section of the outer are
shown in Figure 5 (b), Figures 7 (a) and (b), respectively. Floor framing that forms the office space was
not incorporated into the analysis, because it bears only permanent load.

4-D35
0.7m

Column
1.0m

ST-D16@100
12-D38
4-D35
Fc60 0.7m HOOP-D13@100
Fc36 0.7m
mark:X-shaped bar Steel beam H-6002001219
(a)Cross-section of the boundary beam (b)Cross-section of outer frame
Figure 7. A representative cross-section

Input seismic motion was referred to the figure for the Building-Center-of-Japan standard artificial
earthquake (BCJ-L2). This earthquake was larger than the Government standard artificial earthquake. The
Government standard artificial earthquake is currently applied for this size of building in Japan in seismic
design against large earthquakes. Figure 8 shows the acceleration response spectra of the BCJ-L2,
representative Government standard artificial earthquake (G-art), El centro (NS) and Taft (EW) with the
maximum velocity standardized at 50 cm/sec.

2000
h=0.03

Acceleration(cm/s/s)
1500 El centro(NS)
Taft(EW)
1000
G-art
500 BCJ-L2

0
0 1 2 3 4 5
Period(s)
Figure 8. Acceleration response spectra

Analysis model
The analysis model is shown in Figure 9. A plane analysis model was employed. Columns and beams of
the outer frames, core walls and the boundary beams were modeled with linear members. A rigid floor
was assumed and mass was concentrated at the floor location at each story.
Steel beam Boundary beam

Panel Rigid zone

Reinforced Core wall


concrete column
Rigid beam

Outer frames Core wall frames


Figure 9. Analysis model

A bending skeleton curve of the core wall was modeled tri-linearly from the bending moment(:M)
-curvature (:) relationship based on the Navier hypothesized section analysis [2]. The hysteresis of its
bending skeleton curve was determined to be asymmetrical Mutos loop [3]. The compression and tension
of the axial force during the earthquake were taken into account. The M- relationship of the boundary
beam tying the core walls was modeled tri-linearly with Suganos equation [4]. The hysteresis of its
bending skeleton curve was made as a Takedas loop [5], and a rigid zone was incorporated into the space
between the beam end and the center of gravity of the wall. Individual parts of the outer structure were
modeled as follows. The M- relationship of the column was modeled tri-linearly taking into account
the axial force during the earthquake. The axial force was due to the shear force at the beam yield.
Hysteresis was made to be a Takedas loop. The M- relationship of the steel beam was modeled
bi-linearly by setting the bending yield as an inflection point. Its hysteresis was set to normal bi-linear.
The panel took into account the shear stiffness by elasticity. The shear characteristics of the all members
were set as elastic. The columns and core walls were fixed at the base of each member on the first floor
where the earthquake motion was input. Internal viscous damping is assumed for the vibration model, in
which damping factor is 3% for the instantaneous first natural period.

Analysis results
The natural period and analysis results of the Y direction are shown in Table 1 and Figure 10, respectively.
The story drift angle shown in Figure 10 (b) was under 1/100, which was one of the design criteria. As
shown in Figure 10 (c), the ratio of the shear force that the core wall bears was about 95% and 85% for
the lower and middle stories, respectively. It is found that most of the seismic force was born by the core
wall.

Table1. Natural periods


Mode number X-direction Y-direction
1st natural period(s) 2.56 1.97
nd
2 natural period(s) 0.62 0.57

25 25 25

20 20 20 Total

15 15 15
Story

Story

Story
10 10 10

5 5 5
Outer frame
0 0 0
0 100 200 300 400 0 0.005 0.01 0.015 0E+0 2E+4 4E+4 6E+4
Acceleration(cm/s/s) Story drift angle(rad) Shear force(kN)
(a) Maximum acceleration (b) Maximum story drift angle (c) Maximum shear force
Figure 10. Analysis results of the Y direction (BCJ-L2 input)

25 25 25

20 20 20

15 15 15
Story

Story

Story

My Su
10 10 10
M
Sc
5 5 5
S
0 0 0
0E+0 1E+5 2E+5 3E+5 0 1 2 3 4 5 0 0.01 0.02 0.03
Bending moment(kNm) Shear stress(N/mm/mm) Rotation angle(rad)
(a) Maximum bending moment (b) Maximum shear stress (c) Maximum Rotation angle
Figure 11. Analysis results of the Y direction (BCJ-L2 input)

The stress of individual core wall (W2 in Figure 1) of the subject building and the rotation angle of the
members (B2 in Figure 1) of the boundary beam are shown in Figure 11. The representative bending
moment (:M) of the wall shown in Figure 11 (a) was lower than the yield bending moment (:My) of the
wall. The shear stress (:S) in the wall shown in Figure 11 (b), for which the maximum value at the base of
wall was about 2 N/mm2. It was lower than the shear crack stress (:Sc) and the ultimate shear strength
(:Su). The rotation angle of the boundary beam shown in Figures 11 (c) was about 1/40. These results
were specified as conditions of the loading test described in the next section.
An analysis model was constructed to review the effects of the boundary beam. An elastic boundary
beam having an equivalent stiffness to the boundary beams stiffness during an earthquake was
incorporated into the model. Elastic characteristics were provided by setting the bending stiffness of the
boundary beam to 1/50. Thus, the natural period during an earthquake was almost identical, and the
boundary beam did not dissipate energy in the analysis model. The response of this analysis model
(:Comp model) and the response of the HMT were compared at X direction, as shown in Figure 12. It was
confirmed that the response largely decreased due to the vibration energy dissipation by the boundary
beam.

25 25 25
HMT
20 20 20

15 15 15

Story
Story

Story

Comp model Comp model


10 10 10

5 5 5
Comp model HMT HMT
0 0 0
0 0.005 0.01 0.015 0E+0 1E+5 2E+5 3E+5 0 1 2 3 4
Story drift angle(rad) Bending moment(kNm) Shear stress(N/mm/mm)
(a) Maximum story drift angle (b) Maximum bending moment (c) Maximum shear stress
Figure 12. Analysis results of the X direction (BCJ-L2 input)

LOADING TEST OF EARTHQUAKE RESISTING WALLS WITH BUILD-IN STEEL

Test purpose
The test was conducted to confirm that (1) the shear strength of the earthquake-resisting wall with built-in
steel exceeded that of a reinforced concrete earthquake-resisting wall, and (2) the bending skeleton curve
could be represented by the design equations of a reinforced concrete earthquake-resisting wall where the
bending reinforcement was replaced with a steel frame. It is thus confirmed that the available study
results for the reinforced concrete earthquake-resisting walls accumulated to present can be utilized in
design.

Test specimen and test method


The configurations and sizes of the test specimens are shown in Figure 13. The test specimens were for
the walls of lower stories of the model building shown in Figure 1 and Figure 7(a). The scale was about
1/5. Bearing plates were welded to the steel frame for connecting to the concrete. Vertical bar of wall and
steel frames were anchored to stubs.
The test specimens were placed in the loading apparatus shown in Figure 14. Tests were carried out with
static cyclic loading. Two kinds of tests were carried out: shear test (WS) and bending test (WB). To
simulate the individual conditions, shearing destruction and bending destruction were induced by
adjusting the ratio of bending moment / shear (:M/Q) in the testing apparatus. M/(QL) of the WS
specimen was 0.67, and it of the WB specimen was 2.0. The axial force was set to have an axial load /
axial strength ratio of 0.2. The material characteristics are shown in Table 2, 3. The concrete had higher
strength than the design strength of 60 N/mm2. It was 68.3 through 83.9 N/mm2 for material ages of 41
through 110 days. Thus, the estimated concrete strength for the age at the tests was 79.9 N/mm2 and 83.0
N/mm2 for WS and WB, respectively. The yield point of the steel frame was higher than the design value
of 380 N/mm2.

Steel frame Vertical bar Lateral bar

0.6m 0.95m 0.5m


BH-601001212 2-D6@145 2-D6@50 Stub Stub
1.9m 1.0m
0.14m

Specimen

L=1.2m Stub
Bearing plate C.L. Stub 0.14m
2-D6@50
2-FB-92548@100
L/2=1.2m/2 3.2m 1.2m
(a)Section of the specimen (b)View of the specimen
Figure 13. Configurations and sizes of the test specimens

Vertical jack3 Reactor


push:2MN frame Horizontal jack2
pull :1MN push:1MN
pull :0.5MN

Displacement:D
131. 98

Load cell:F 600 600

Reactor frame
500

1900
1200

950

Load cell:F
1800

600

3200

Figure 14. Loading apparatus


Kaj i ma

Table2. Mechanical characteristics of concrete


Design Compressive Strain at compressive Youngs Poissons
Age
strength strength Strength modulus ratio
days N/mm2 10-6 103N/mm2
41 68.3 --- --- ---
Fc60 81 79.9 3217 30.2 0.204
110 83.9 3356 30.2 0.181

Table3. Mechanical characteristics of steel


Kind thickness Yield Tensile Strain Youngs Part
point strength at yield modulus
2 2
mm N/mm N/mm 10-6 105N/mm2
D6(SD345) --- 362 527 4137 1.700 Reinforcing bar
PL12(SM490A) 11.67 407 572 1979 2.058 Build-in steel

Test results
The load displacement relationship for the shear test is shown in Figure 15(a). This figure also shows
the calculated ultimate shear strength (:Qu) based on reference [6] in accordance with the method for
reinforced concrete earthquake-resisting walls. The measured maximum strength from the tests was
higher than the calculated value. It was confirmed that the estimate of ultimate shear strength assuming
the relevant wall as a reinforced concrete earthquake-resisting wall was usable for safe design.

3000 1500
2000 1000
Qu=1564kN
Load:F(kN)

Load:F(kN)
1000 500
0 0
-1000 -500
Calculation
-2000 -1000
Experiment
-3000 -1500
-20 -10 0 10 20 30 -30
-20 -10 0 10 20 30
Displacement:D(mm) Displacement:D(mm)
(a) Shear test (b) Bending test
Figure 15. Load displacement relationship

The loading displacement relationship for the bending test is shown in Figure 15(b). This figure also
shows the calculated load displacement relationship. The analysis model used in the calculation, and
settings for the M- relationship and shear (:Q)-displacement (:) relationship, are shown in Figure
16. The same method was used here as for the characteristics settings for the analysis model shown in
Figure 9. Rotation spring elements due to the extracted steel frames evaluated from the test results were
incorporated into the top and base of the wall. Calculation results and test results agreed well. It was
confirmed that the bending skeleton curve used for the HMT could be represented by the same evaluation
method for reinforced-concrete earthquake-resisting walls.
M(kNm)
Bending moment by axial load
2107 EI0.01
941 EI0.438
Load:F
Displacement:D E=2.98104N/mm2
Rigid zone I=3.41106cm4
EI
(Stub part)
Rotational spring Q(kN
Bending shear element
2.64108kNcm/rad (6-division) 955 GA0.116
(M>569kNm) GA0.175
707
G=1.24104N/mm 2
A=1680cm2
GA =1.2

Figure 16. Analysis model and settings for the bending and shear skeleton curve

LOADING TEST OF BOUNDARY BEAM

Test purpose
The purpose was to clarify the energy dissipation capability of X-shaped steel bars and the effect of
reducing damage by using PVA-ECC in the boundary beams. Furthermore, variation of characteristics by
introducing the pre-stress was confirmed. The tests confirmed the characteristics mainly up to the rotation
angle of the member during a large earthquake: 0.03rad.
Test specimen and test method
The configurations and sizes of the test specimens are shown in Figure 17. Table 4 summarizes the test
specimens. The test specimens were for the boundary beams of the each story of the model building. The
scale was 1/2. Three kinds of test specimens were used: (a) normal concrete with a standard bar
arrangement (NC), (b) PVA-ECC with X-shaped steel bars (FC-X), and (c) FC-X where pre-stress was
introduced (FC-X-P). The reinforcing bars in each test specimen were adjusted to make the bending
strengths almost identical. Reinforcing bars and pre-stress bar were anchored to stub.
The test specimens were placed in the loading apparatus shown in Figure 18. A static cyclic loading
was carried out to accomplish inversed symmetry bending. The material characteristics are shown in
Table 5, 6.
0.5m

0.5m

0.5m
Pre-stress
mark
0.35m 0.35m :X-shaped bar 0.35m
(a)Cross-section of NC (b)Cross-section of FC-X (c)Cross-section of FC-X-P
Figure 17. Configurations and sizes of the test specimens

Table 4. Test specimen of boundary beams


Name of specimen NC FC-X FC-X-P
Concrete normal PVA-ECC PVA-ECC
Horizontal bar 4-D19 2-D16 2-D13
Reinforcing bar X-shaped bar --- 2-D16 2-D13
Pt(%) 0.73 0.51 0.32
Reinforcement 2-D10@75
Stirrup
Pw(%) 0.54
Diameter(mm) --- --- 32
Pre-stress
Force(kN) --- --- 580

7950
750 2500 2750 1950
Displacement:D
700

Horizontal jack

Stub push:2MN

pull :1MN Reactor
750


L=1.5m
3600

wall
2600

750

Specimen Load cell:F


Stub
Vertical jack push:2MN
pull :1MN
300

Figure 18. Loading apparatus


Table5. Mechanical characteristics of concrete
Name of Compressive Youngs Strain at maximum Tensile strength
specimen strength modulus strength
2 4 2 2
N/mm 10 N/mm N/mm
NC 36.1 2.92 1969 2.99
FC-X 35.6 1.52 3850 4.73
FC-X-P 38.7 1.83 3414 4.34

Table6. Mechanical characteristics of steel


Kind Yield point Tensile Strain at Youngs Part
strength yield modulus
N/mm2 N/mm2 10-6 5
10 N/mm
2

D10(SD295) 360 495 1870 1.93 Stirrup


D13(SD390) 409 586 2290 1.78 Reinforcing bar
D16(SD390) 440 610 2350 1.87 Reinforcing bar
D19(SD390) 443 612 2260 1.96 Reinforcing bar

Test results
The load displacement relationship is shown in Figure 19. The cracking conditions are shown in Figure
20. The strength of the NC started decreasing at a member rotation angle (:D/L) of about 1.5%. Slip
characteristics were dominant in the hysteresis loop. However, the strength of the FC-X generally did not
decrease up to the target rotation angle 0.03 of the member. The near spindle-shaped hysteresis loop was
also observed. The hysteresis loop of the FC-X-P resembled the origin-oriented type due to the effects of
the pre-stress. The cracking conditions of FC-X-P and FC-X were described as slight compared with
those of the NC.

400 400 400

200 200 200


Load:F(kN)

Load:F(kN)

Load:F(kN)

0 0 0

-200 -200 -200

-400 -400 -400


-50 -25 0 25 50 -50 -25 0 25 50 -50 -25 0 25 50
Displacement:D(mm) Displacement:D(mm) Displacement:D(mm)
(a) NC (b) FC-X (c) FC-X-P
Figure19. Load displacement relationship

(a) NC (b) FC-X (c) FC-X-P


Figure 20. Cracking conditions

The equivalent damping factor (:heq) calculated from the test results is shown in Figure 21. The
calculation method is shown in Figure 21. The equivalent damping factor of the FC-X was the largest
throughout almost all the member rotation angles. Accordingly, the boundary beam using X-shaped steel
bars and PVA-ECC provided good energy dissipation performance and decreased damage.
Equivalent damping factor

0.25
FC-X
1 W
0.2 heq =
FC-X-P W 2 We
0.15 Pmax 1
We = Pmax max
0.1 2
NC
0.05

0 max
0 0.5 1 1.5 2 2.5 3
Rotation angle: D/L(%)
Figure 21. Equivalent damping factor calculated from the test results

CONCLUSIONS

The HMT proposed in this study is an advanced structural framing system that has many merits in
planning and facilities. HMT is a hybrid structure using earthquake-resisting core walls, outer frames and
inversed steel haunch beams. Also, it has seismic response control performance by combining core walls
and boundary beams. As a result, HMT economically achieves high structural safety against large
earthquakes. Furthermore, manufacturing efficiency is greatly increased by unifying the members of the
outer frames and the boundary beams, the configuration and bar arrangement of the core walls, and the
member of the floor framings that form the office areas. Utilizing the above-mentioned features, the
structural construction cost is reduced by 15% compared to the concrete filled steel tube structure, and
short-term construction is achievable. The seismic response analysis results illustrated that the
earthquake-resistance standards of Japan, as a severely seismic country, could be satisfied, and the
boundary beams reduced the seismic response. The loading tests confirmed that the shear strength and
bending skeleton curves of the earthquake-resisting walls with built-in steel could be evaluated by
conventional design equations for reinforced concrete earthquake-resisting walls. It was also verified that
the boundary beams as dampers had a large equivalent damping factor and could decrease damage
compared with boundary beams of normal cross-sections.

ACKNOWLEDGEMENT

Great cooperation was provided to this study by Akita Prefectural University and Graduate School
of Architecture and Building Science, Tohoku University. The tests for the earthquake-resisting wall
with built-in steel were conducted by Professor J. Kobayashi, Assistant professor T. Nishida,
Research associate N. Teramoto of Akita Prefectural University. The tests for the boundary
beams were conducted by Associate professor M. Maeda, Research associate N. Hori of Tohoku
University. The authors express their sincere appreciation.

REFERENCES

1. Kanda T., Material Design Technology for High Performance Fiber Reinforced Cementitios
Composite, Concrete Journal, Vol.38, No.6, 2000: 9-16 (in Japanese)
2. S. A. Kaba, S. A. Mahin, Refined Modelling of Reinforced Concrete Columns for Seismic
Analysis, University of California:Berkeley, California, Report No. UCB/EERC-84/03, 1984.4
3. Muto K., Hisada T., Tsugawa T., Bessho S., Eartkquake Resistant Design of a 20 Story Reinforced
Concrete Building, Proceedings of 5th World Conference on Earthquake Engineering, Roma, Italy,
1972: 1960-1969
4. Architectural Institute of Japan, Standard for Structural Calculation of Reinforced Concrete
Structures-Based on Allowable Stress Concept-, 1999: 56-59 (in Japanese)
5. Takeda T., M.A. Sozen, N.N. Nielsen, Reinforced Concrete Response to Simulated Earthquakes,
Proceedings of American Society of Civil Engineers, Vol.96, No. ST12, Dec. 1970: 2557-2573
6. Architectural Institute of Japan, Design Guidelines for Earthquake Resistant Reinforced Concrete
Buildings Based on Inelastic Displacement Concept, 1999: 208-240 (in Japanese)