Structures Congress 2015 2792

Seismic Design and Analysis of a 20-Storey Demonstration Wood Building

Zhiyong Chen1; Ying Hei Chui2; and Marjan Popovski, M. ASCE3

1
Ph.D., Wood Science and Technology Centre & Department of Civil Engineering,
University of New Brunswick, P.O. Box 4400, Fredericton, NB, Canada E3B 5A3.
E-mail: [email protected]
2
Ph.D., P.Eng., Wood Science and Technology Centre, University of New
Brunswick, P.O. Box 4400, Fredericton, NB, Canada E3B 5A3. E-mail:
[email protected]
3
Ph.D., P.Eng., Advanced Building Systems, FPInnovations, 2665 East Mall,
Vancouver, BC, Canada V6T 1W5. E-mail: [email protected]

Abstract

This paper presents the seismic design and analysis of a 20-storey

demonstration wood building, which was conducted as a part of the NEWBuildS tall
wood building design project. A hybrid lateral load resisting system was chosen for
the building. The system consisted of shear walls and a shear core, both made of
structural composite lumber, connected with dowel-type connections and heavy-duty
HSK (wood-steel-composite) system. The core and the shear walls were linked with
horizontal steel beams at each floor. The wood-based panel-to-panel interface was
designed to be the main energy dissipating mechanism of the system. A detailed
finite element model of this building was developed and non-linear time history
analyses were performed using 10 earthquake motions. The results showed that the
seismic response of the 20-storey demonstration building met the various design
criteria and the design details are appropriate.

INTRODUCTION

Wood as a structural material dates back more than 7000 years. The ancient
construction techniques of using wood evolved over time and reached the peak
between the 6th and the 9th century, when several tall or large buildings were
constructed. Currently the tallest ancient wood building in the world is the Pagoda of
Fogong Temple (commonly known as Yingxian Wood Pagoda) in Shanxi, China,
with nine storeys and 67.31m in height, that was built in 1056 AD (Chen 2011).
In the past century, with the introduction of building codes around the world,
wood structures are usually restricted to a maximum of 3-6 storeys primarily due to
concerns about fire safety. Nowadays, however, wood is attracting a lot of global
attention for use in mid- and high-rise buildings, due to its low carbon footprint and
fast construction time. Consequently, several contemporary high-rise residential
wood buildings were constructed around the world. For example, the tallest
residential wood building in the world currently is the Forté Building in Melbourne,
Australia which is 10 storeys high with the first storey constructed with reinforced
concrete and the upper 9 storeys constructed with Cross Laminated Timber (CLT).

© ASCE
Structures Congress 2015 2793

As global intereest in using wood produucts in tall buuildings intennsifies due to

t the good
envvironmental credentials of o wood, it is
i expected that
t more taall wood buildings will
be designed
d andd constructed in the com
ming years.
Wood products
p possess a high strength/weiight ratio sim milar to steeel, and as a
resuult, they coould be suittable for constructing high-rise
h buuildings. Thee practical
heigght limit foor a wood building is governed by b several factors, including the
perrformance of o the struuctural woood products, connectiions and joints, the
connstruction teechnology, and
a design methods.
m A tall wood building desiign project
wass initiated ass a part of thhe NEWBuiildS (NSERC C Strategic Network
N on Innovative
Woood Productss and Buildiing Systemss) research network.
n A team of NE EWBuildS
reseearchers in collaboratioon with desiggners and architects
a connducted a design
d of a
20--storey demo onstration wood
w buildinng known ass the CHEC CKER buildiing. Fig. 1
shoows a model of the CHE ECK buildingg. The goal ofo the projecct was to shoow that tall
woood buildingss can be a viable alternattive when laatest wood prroducts and innovative
connnection sysstems are used,
u and too identify potential techhnical challlenges and
corrresponding solutions foor tall wood buildings. This
T paper focuses
f on thhe seismic
dessign and anallysis of this demonstration building.

Fig. 1. The 20-stoorey demonstration CH

HECKER wood
w buildin
ng

© ASCE
Structures Congress 2015 2794

STRUCTURAL DESIGN

Input Data for the Design

It was assumed that the 20-storey demonstration building is located in North
Vancouver, BC, Canada, a location that has high earthquake, wind and rain loads.
The tall wood building is composed of upper 19-storey wood construction and a
bottom-storey concrete podium. The total building height is 60m (3m per storey),
and the standard plan dimensions are 27m × 27m with a 9m frame gird. The
importance category of “Normal” and a site class of “D” were used for the seismic
design. Various parameters linked to the design of the building are shown in Table 1
and were based on the 2010 National Building Code of Canada [NBCC] (NRC 2010)
for the North Vancouver location.

Table 1. Design data for North Vancouver

Load Item Value Load Item Value
a
Roof 3.00 kPa IW 1.0 (ULS), 0.75 (ULS)
b Wind
Dead Floor 5.00 kPa q 0.35 kPa (1/10), 0.45 kPa (1/50)
Partition 0.50 kPa IE 1.00 (ULS)
Roof 1.82 kPa Sa(0.2) 0.88
Live
Floor 1.90 kPa Sa(0.5) 0.61
Earthquake
IS 1.0 (ULS), 0.9 (ULS) Sa(1.0) 0.33
Snow SS 3.00 kPa Sa(2.0) 0.17
Sr 0.30 kPa PGA 0.44 g

Note: a – including allowance for roof top units and screens; b – including 125mm concrete topping.

Design Aspects for the Building

The conceptual design was chosen to lead to the development of a safe and
economical structure that can provide the optimum performance under various loads.
Since the concrete podium has much higher lateral stiffness than the wood
superstructure, the structural design and analysis of this building due to seismic
loads focuses on the wood portion of the structure only.
For the 19-storeys wood structure, a combination of laminated strand lumber
(LSL) shear walls with a LSL core was chosen as the lateral load resisting system
(LLRS) for the building. These two subsystems, the shear walls and the core, were
linked by steel beams with hinge joints, as illustrated in Fig. 2.
To reduce the number of horizontal connections between panels and
minimize the vertical deformation induced by the compressive stresses perpendicular
to grain on the wood floor assemblies, a balloon framing technique using LSL panels
with a length of 19m was selected for the shear walls and the core. Since the length
of the shear walls and the core walls was larger than the production width of the

© ASCE
Structures Congress 2015 2795

panels (2.44m for LSL), vertical joints (Fig. 2) were needed to connect the adjacent
panels. To resist the shear between panels or between panel and the podium, as well
as the overturning uplift force, shear connectors and hold-downs were required, as
shown in Fig. 2.

Fig. 2. Schematic diagram of the LLRS

The mechanical connections in mass timber structures are the weakest link
and source of ductility of the whole structural system, so the selection of the
connection type, its failure mode characteristics and the sequence of failure among
the connections affect the structural performance of the buildings under seismic
actions. In an attempt to design this tall wood building with higher stiffness, strength
and ductility, dowel-type connections and HSK (Wood–Steel–Composite) system
were used in the vertical joints of the shear walls and the core, respectively. The
HSK system is also used as the shear connector and hold-down connections. The
vertical joints of the shear walls (dowel-type connections) and the core (HSK system)
were assumed to yield sequentially, and the ultimate limit state of the building is
defined as the failure of the shear connectors and hold-downs.
Since the structural system and the connections are novel, no design
provisions are available in NBCC (NRC 2010), and the Canadian Standard for
Engineering Design in Wood, CSAO86 (CSA 2014). Therefore principles of
mechanics and numerical simulation were used to design and estimate the structural

© ASCE
Structures Congress 2015 2796

responses of the assemblies, connections and the whole building. Based on the
gravity, wind and seismic loads and considering the fire performance of the
structural members, the plan and elevation of the final design of the building are
shown in Figs. 3 and 4 and the details are discussed below.

Fig. 3. Typical floor plan of the building

Grade 2.1E “TimberStrand○R LSL” (CCMC 1994) with dimensions of 19m ×

2.44m × 89mm (length × width × thickness) was used to build the shear walls and
the core. Three layers of LSL panels were combined together to achieve a total wall
thickness of 267mm. Steel beam S5×10 of Grade 50 (CSA 2009) was used to link
the shear wall to the core sub-system (Figs. 2 and 3). Dowel-type connections in the
LSL were used in the vertical joints of the shear walls, while the HSK system was
used for the vertical joints of the core, shear connectors, and hold-downs for the
whole system. For the dowel-type connection, Pa11A (Moses 2000), the diameter of
dowel is 19mm, and the stiffness and strength of each dowel were 25.5kN/mm and
32.5kN, respectively. The HSK system is a patented product of TiComTec GmbH,
and the performance parameters were provided by Prof. Bathon from the University
of Wiesbaden, Germany. The stiffness and strength of each “hole” of the HSK
system parallel to the grain were 7.4kN/mm and 0.8kN; while those in the

© ASCE
Structures Congress 2015 2797

perpendicular direction were 2.5kN/mm and 0.8kN respectively. Assumed design

value for all connections was taken as the mean test strength divided by a factor of
1.6.

(a) South (b) West

Fig. 4. Elevations of the building

NUMERICAL SIMULATIONS

Finite Element Model Development

Since the LLRSs in the two orthogonal directions are nearly identical (Fig. 2),
only the LLRS in the east-west direction was analyzed. A two-dimensional (2D)
modeling approach was adopted for the analyses and a 2-D finite element model
(FEM) of the tall wood building (Fig. 5) was developed in ABAQUS (Hibbitt et al.
2013). As shown in Fig. 5, the LSL panels were connected together by vertical joints,
shear connectors and hold-downs. The panels were attached to the concrete podium
using shear connectors and hold-downs. The openings for doorways were
punched-out from the LSL panels. Steel beams were used to link the shear walls to

© ASCE
Structures Congress 2015 2798

the core, and to transfer the lateral load. The connections between steel beams and
LSL panels were simulated by hinge joints.

Force

Deformation

(a) Vertical & shear connectors

Force

Deformation

(b) Hold-down connector

Fig. 5. 2-D FEM of the building Fig. 6. Macro-element connectors

According to capacity based seismic design, in timber buildings the

connections are designed to be the weaker part of the system. Consequently, this
demonstration building can be regarded as a structural system consisting of strong
wood-based assemblies and weak connections. Thus, the structural assemblies and
connections were modeled with linear and non-linear behavior, respectively. The
constitutive models for the LSL panels and steel beams were assumed to have
orthotropic and isotropic elasticity, respectively. The properties of these two
materials used in the FEM are shown in Table 2.

Table 2. Material properties for steel and LSL

Material E1, MPa E2, MPa v12 G12, MPa
LSL 14,480 810 0.741 905
Steel 206,000 206,000 0.3 79,230

© ASCE
Structures Congress 2015 2799

Because the load-deformation response data of HSK system and specific LSL
dowel connection were unavailable, idealized models (Fig. 6) were used to simulate
the reversed cyclic response of the connections. The behavior of the vertical joints
and the load-deformation response of the shear connectors were assumed to be
idealized elasto-plastic with isotropic hardening (Fig. 6a). In order to assign a
specific ductility ratio to these connections, damage behavior was employed by the
connection models with a motion-based linear damage evolution law. Meanwhile,
the element deletion technique as used to delete a connection from the mesh once it
was damaged completely. The hold-down connection was assumed to behave similar
to the shear connectors and vertical joints under tension, but resist compression
without deformation (Fig. 6b), since most of the compression deformation happens
by bearing in LSL and very small deformation occurs in HSK system. The
parameters of these connection models are shown in Table 3. As the behavior of the
connections between panels and between panel and the concrete podium was
simulated by the connection models above, the interaction relationship between
connecting components was simplified as frictionless contact in tangential direction
and hard contact in normal direction.

Table 3. Connection model parameters

Connection Storey K, Pmax,
Core/Wall Connector Δ y , mm Δ max , mm μ
Type No kN/mm kN
Core HSKPar 1,780 192 ±0.1 ±0.5 5
1
Wall Dowel 67 100 ±1.5 ±3.0 2
Core HSKPar 1,780 192 ±0.1 ±0.5 5
Vertical Joint 2
Wall Dowel 57 85 ±1.5 ±3.0 2
Core HSKPar 1,180 128 ±0.1 ±0.5 5
3
Wall Dowel 40 60 ±1.5 ±3.0 2
Core 600 192
1
Wall 300 96
Shear Core 600 192
2 HSKPer ±0.3 ±1.6 5
Connector Wall 300 96
Core 400 128
3
Wall 200 64
Core 50,000 5,410
1
Wall 25,000 2,700
Core 27,540 2,980
Hold-Down 2 HSKPar +0.1 +0.5 5
Wall 13,770 1,490
Core 9,440 1,020
3
Wall 4,720 510

Note: ‘HSKPar’ and ‘HSKPer’ indicate the major direction of HSK connection parallel and
perpendicular to LSL grain.

© ASCE
Structures Congress 2015 2800

The shear core and shear walls were meshed using a 4-node bilinear plane
stress element (CPS4R) with reduced integration and hourglass control. The steel
beams were modeled using a 2-node linear Timoshenko (shear flexible) beam
element (B21). A 2D, 2-node connector element (CONN2D2) was used to simulate
the vertical joints, shear connectors and hold-downs. Point mass element (MASS)
was used to assign the corresponding mass to each storey at the floor or roof level.
The FEM was meshed with a total of about 30,000 elements, including the shell,
beam, connector and mass elements.
Non-linear Time History Analysis
A series of comprehensive numerical simulations, including frequency,
gravity-induced response, response spectrum, pushover, and non-linear time history
analysis, were performed to check the performance of the building as part of the
NEWBuildS tall wood building design project. Only results from the non-linear time
history analyses are presented in this paper.
The seismic behavior of the tall wood building model under design hazard
level was analyzed with the explicit dynamic analysis method using the
central-difference operator (Hibbitt et al. 2013). Ten “Far-Field” earthquake records
(NGA #: 169, 721, 752, 829, 900, 1111, 1158, 1244, 1602, 1787) in the fault normal
(FN) direction (ATC 2009) were scaled at the corresponding fundamental period of
the building model (1.97s) to match the spectral acceleration, Sa, of the North
Vancouver design spectrum, as shown in Fig. 7.
10

1
Spectral Acceleration, Sa(g)

0.1

0.01

Target Spectrum
Results Geom. Mean
1E-3
0.01 0.1 1 10
Period, T(s)

Fig. 7. Scaled earthquake records used in the analyses (To=1.97 s)

Fig. 8 shows the scaled input acceleration of an earthquake (NGA #: 1602)

with a duration of 15s, as well as the structural responses, in terms of base shear,
lateral drift at top and inter-storey drift ratio of the building.

© ASCE
Structures Congress 2015 2801

0.50 6000

0.25 3000
Acceleration, g

Base shear, kN
0.00 0

-0.25 -3000

-0.50 -6000
0 3 6 9 12 15 0 3 6 9 12 15
Time, s Time, s

(a) Scaled earthquake excitation (b) Base shear

200

100
Lateral drift at top, mm

-100

-200
0 3 6 9 12 15
Time, s

(c) Lateral drift at top (d) Inter-storey drift ratio

Fig. 8. Earthquake excitation NGA # 1602 (a) and FEM responses: (b) base
shear, (c) lateral drift and (d) inter-storey drift ratio

The statistics for the base shear and inter-story drift ratio of the tall wood
building under the ten earthquake excitations are shown in Fig. 9. Also shown in Fig
9 is the design base shear of 4893kN estimated by Equivalent Static Force Procedure
of NBCC (NRC 2010), the yield load of 7430kN calculated using the Equivalent
Energy Elastic-Perfectly-Plastic (EEEP) procedure of ASTM 2126 (ASTM 2009),
and the maximum capacity of 8140kN of the building calculated from Pushover
Analysis. As illustrated in Fig. 9a, most (nine) of the base shears of the building are
higher than the design value of 4893kN. However, all the base shears are less than
the yield load of 7430kN and the maximum capacity of 8140kN. In addition, as
shown in Fig 9b, all the inter-storey drift ratios are less than the design criterion of
2.5% (NRC 2010). All results show that the design of the building was adequate. It
should be noted that under all earthquakes only the vertical joints between the shear
walls of the tall wood building showed yielding behavior.

© ASCE
Structures Congress 2015 2802

(a) Base shear

(b) Inter-storey drift ratio

Fig. 9. Seismic response of the tall wood building

SUMMARY & CONCLUSIONS

This paper presents the seismic design and analysis of a 20-storey

demonstration wood building, which was conducted as a part of the NEWBuildS tall
wood building design project in Canada. A hybrid lateral load resisting system was
chosen consisting of LSL shear walls and LSL shear core, connected with
dowel-type connections and heavy-duty HSK (wood-steel-composite) system. The
core and the shear walls were linked with horizontal steel beams at each floor. The
2D finite element modeling approach using the concept of strong assembly and weak
connection was appropriate to investigate the seismic response of the structural and
the connection system of the 20-storey wood building. Results from the analyses
showed that the hybrid system was suitable for the 20-storey high-rise wood building
located in North Vancouver, BC. The system provided sufficient stiffness, strength

© ASCE
Structures Congress 2015 2803

and ductility to the building.

ACKNOWLEDGEMENTS

The authors greatly acknowledge the financial support provided by Natural

Sciences and Engineering Research Council (NSERC) of Canada under the Strategic
Research Network on Innovative Wood Products and Building Systems
(NEWBuildS) and Forestry Innovation Investment Ltd (FII), BC, Canada. Thanks
are also extended to Mr. Eric Karsh and Dr. Mahmoud Rezai from Equilibrium
Consulting Inc., Vancouver, BC, and Mr. Mark McCormick from exp Services Inc.,
Fredericton, NB, for providing technical advice, and Dr. Leander Bathon from
University of Wiesbaden, Germany for providing the mechanical specifications and
test data on the HSK connection system.

REFERENCES

ASTM. (2009). Standard test methods for cyclic (reversed) load test for shear
resistance of framed walls for buildings. ASTM E2126. American Society for
Testing and Materials, West Conshohocken, PA.
ATC. (2009). Quantification of building seismic performance factors.
ATC-63/FEMA P-695 Project Rep., Applied Technology Council, Redwood City,
CA.
CCMC. (1994). TimberStrand○R LSL. Evaluation Report CCMC 12627-R. Canadian
Construction and Materials Centre, Ottawa, ON.
CSA. (2014). Engineering Design in Wood. CSA O86-14. Canadian Standards
Association, Toronto, ON.
CSA. (2009). Design of Steel Structures. CSA S16-09. Canadian Standards
Association, Toronto, ON.
Chen, Z. Y. (2011). Behaviour of typical joints and the structure of Yingxian Wood
Pagoda. Ph.D. dissertation, School of Civil Engineering, Harbin Institute of
Technology, Harbin.
Hibbitt, D., Karlsson, B., and Sorensen, P. (2011). ABAQUS Analysis User’s Manual
(Version 6.11). Dassault Systemes Simulia Corp., Pawtucker, RI.
Moses, D.M. (2000). Constitutive and analytical models for structural composite
lumber with applications to bolted connections. Ph.D. dissertation, Department
of Civil Engineering, University of British Columbia, Vancouver, BC.
NRC. (2010). National Building Code of Canada, Canadian Commission on
Building and Fire Codes, National Research Council, Ottawa, ON.
TiComTec GmbH. (2011). Load Bearing Constructions using Wood-Concrete-
Composite Technique with Glued-in HBVR – Shear Connectors. TiComTec
GmbH Technical Report, Haibach.

© ASCE