shaking table tests at national laboratory for civil

Seismic Engineering Research Infrastructures for European Synergies

Investigation of seismic performance of multi-storey timber buildings within the framework of the SERIES Project

Shaking table tests at National Laboratory for Civil Engineering (Lisbon Portugal)

M. Piazza – University of Trento

A. Campos Costa – LNEC (Lisbon)

P. Candeias – LNEC (Lisbon)

1

SERIES Concluding Workshop Ispra 28-30 May 2013

Introduction - Constructive systems - Material and geometry - Tests setup - Accelerograms – Test 2

• Introduction

• Constructive systems

• Material and geometry

• Tests setup

• Accelerograms – test procedure

• Tests

• Preliminary results



FACILITY: NESDE shake table, LNEC, Lisbon (PT)

TA AGREEMENT : January 2010 - Grant agreement n° 227887

STARTING DATE:

16th July 2011 kick-off meeting

END DATE:

21st February 2013

LEAD USER:

Maurizio Piazza & Roberto Tomasi, University of Trento (IT)

ADDITIONAL USERS:

Gerhard Schickhofer, TU Graz, AT Jorge Branco & Paulo B. Lourenço, University of Minho, PT

University of Trento Daniele Casagrande, Paolo Grossi, Maurizio Piazza, Tiziano Sartori, Roberto Tomasi

TU Graz Gerhard Schickhofer, Georg Flatscher

University of Minho Jorge Branco, Paulo B. Lourenço

LNEC Alfredo Campos Costa, Paulo Xavier Candeias



The research on timber buildings involves University of Trento, Italy as the lead institution, University of Minho, Portugal, University of Graz, Austria.


Introduction - Constructive systems - Material and geometry - Tests setup - Accelerograms – Test

The goal of the research presented is the analysis of the seismic performance of timber buildings, with four kinds of timber housing system

5




6

University of Trento Light framed walls system




7


University of Minho Log house system




8


University of Minho Log house system

University of Graz CLT system



Light timber framed wall construction (TF)



Timber frame Usually in the European countries solid wood

elements 60/100x160mm

Structural sheathing panels commonly OSB – gypsum fiber – plywood are used

Connection between frame and panels Depending on the sheathing panels material

ring nails or staples can be used



Structural sheathing panels commonly OSB – gypsum fiber – plywood are used

Floors/roof Prefabricated box elements or classic

beam+boards/panels



Connection between walls: in order to avoid the uplift and the slip of the walls, different fixing system are used

Moreover the wall are screwed along the corner of the building (box behaviour)

Walls and floors are connected by self tapping screws



Connection between walls and foundation slab

Hold down

Angle bracket Alternative system

Tensile connection

Shear connection



Log house construction (LH)



Inner and outer walls are usually made of double/triple laminated beams placed one above the other

Different geometry of corner joint can be used



Floors and roof are usually assembled with timber beams and panels

To avoid the problems related to the different degrees of wood shrinkage perpendicular to the grain, specific attention is required on the construction details



Connection between timber walls and reinforced concrete substructures

The deformation of the wall in the vertical direction (perpendicular to the grain direction) due to the moisture variation must be taken into account also in the connections

Self tapping screws are often used:

- near the door/window opening

- to connect the upper logs (under the beam of the roof)

- to connect the sill logs



Cross Laminated Timber construction (CLT)



CLT panels can be used as vertical elements (walls) or horizontal elements (floors/roofs) depending on the orientation of the layers

- Thickness can vary from 75 to 300 mm (3 to 7 layers) - Width up to 3 m - Length up to 16 m



Connections

The connection between walls and floors are similar to the ones presented for the LTF

CLT Panel connection



Preliminary tests on the single component



Light timber framed system



Connections



Real scale tests (Walls + connections)



WA_A WA_B WA_C WA_D WA_E

CLT connection + walls



Log house connection and walls



Shaking Table Tests

29



Specimens geometry

All specimens had the same geometry (squared 7 m x 5 m) and architectural layout.

• TF/CLT three storey to a maximum height to the peak of 7.65m

• LH two storey to a maximum height to the peak of 5.28m

All the four tested building were designed in accordance with EC 5/EC 8.

30



In order to guarantee the comparison between the different systems, permanent loads and variable loads were the same (permanent: 1.3 kN/m2 - Variable load: 2 kN/m2)

31

Plan and the opening distribution were designed to reproduce a real single-family home.



Log house specimen The log house specimen has been produced by the Portuguese company Rusticasa. According to the present production standard of the company only two storeys have been built, with a maximum height of 5.28 m at the ridge.

32



Walls Inner walls logs 80x160 mm

Outer walls logs 160x160 mm

Floor

Timber beams 90 x 165 mm + 15 mm OSB sheathing panels, ring nails (2.8 x 60 mm)

Connection between walls and floors is obtained by means of dovetail joints

Roof Solid wood rafters (70 mm x 190 mm), over which OSB panels are nailed

The ridge board has a 120 mm x 200 mm cross-section

Connections The sill logs are connected to the steel plate through M16 bolts, class 8.8

33



Light timber framed wall specimen 1 Two different timber frame specimens were tested. The first one, with OSB structural sheathing panels, has been produced by the Italian company Legnocase. The structure was completed with external and internal claddings and one room of the 2nd storey was also equipped with laminate floor - drywall and ceilings – doors - windows and furnitures.



Walls light frame walls (60/100x160 mm studs and 60 mm top/bottom beams)

OSB sheathing panels

Floor

600 mm x 140 mm modular Timber box elements (beams 78x31 mm upper and lower

boards 31 mm) + 15 mm OSB sheathing panels nailed with ring nails 2.8x60 mm)

Connection between walls and floors is obtained by means of screws

Roof Solid wood beams rafter 100x140 mm/760 mm, ridge beam 160x240 mm, purlins

160x160 mm. Wooden plank (20 mm) reinforced with perforated metal strips

Connections

Shear connections: steel plate (anker nails 4x60 mm)

Uplift connections: tie-downs (anker nails 4x60 mm)

Base shear connections: screws 8x180 mm

Base uplift connections: hold-downs (anker nails 4x60 mm)



Light timber framed wall specimen 2 The second TFS building, produced by Italian company Rubner_haus, was built with gypsum fibre structural sheathing panels connected to the timber frame of the walls by means of steel staples (instead of using the system OSB + ring nails)



Walls light frame walls (80x120/160 mm studs and 60 mm top/bottom beams)

Gypsum fiber sheathing panels

Floor

Timber beams 80x200 mm + 12 mm OSB lower panels and 22 mm OSB upper panels

(all panels nailed with ring nails 2.8x60 mm)

Connection between walls and floors is obtained by means of screws

Roof Solid wood beams rafter 120x160 mm/840 mm , ridge beam 160x240 mm, purlins

160x240mm. Wood planks (20 mm) reinforced with perforated metal strips

Connections

Base shear connections: angle brackets (anker nails 4x60 mm)


Shear connections: angle brackets (anker nails 4x60 mm)

Uplift connections: tie-downs (anker nails 4x60 mm)



Cross laminated timber specimen The last test of Timber Building project within SERIES was carried out on CLT three storey building. In this case all the elements (walls, floors, roof) were built with cross laminated timber panels of different thickness.



Walls 3 layers CLT wall panel 100 mm (layup 30-40-30)

Floor

5 layers CLT floor panel 150mm (layup 30-30-30-30-30). Connection between walls

and floors is obtained by means of screws

Roof 3 layers CLT floor panel 99mm (layup 33-33-33). Connection between walls and roof

panels is obtained by means of screws

Connections

Base shear connections: angle brackets (anker nails 4x60 mm)


Shear connections: angle brackets (anker nails 4x60 mm)

Uplift connections: hold-downs (anker nails 4x60 mm)



To test real scale lightweight timber structures the size of the table (5.6x4.6 m) was

enlarged with a structural base frame (5x7m) made of steel beams bolted on the top plate.

Steel plates, anchored to the floor, reproduces the weight

according the load combination for seismic load cases

(self-weight of flooring and a fraction of variable load ).



Experimental data were recorded for accelerations, forces, absolute/relative

displacements.

Type Qty Measure

LVDT 4 Inter-storey drift

LVDT 8 Wall sliding

LVDT 10 Wall uplift

Accelerometers on

structure 39

Accelerations at

different levels and

Accelerometers on

steel basement 5

Accelerations at the

shaking table level

positions

Load cell 10 Forces on hold down

anchoring elements

Optical displacement

measurement system 1

5

Point absolute

displacements (5 points

x,y component)

Optical displacement

measurement system 2 20

Point absolute

displacements



LVDT displacement transducers

Accelerometers



Load Cell LNEC optical displacement measurement system



The accelerograms were recorded at the station "Ulcinj - Hotel Albatros" located at an

epicentral distance of 21 km during the Montenegro earthquake of 15/04/1979 (Mw 6.9).

The values of peak ground acceleration (PGA) at which the buildings were been tested, at

different stages of tests, ranged from 0.07g to 0.5g.

X-record Y-record

48



0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

Rusticasa building

0.28g

0.07g

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.50g

0.15g

This stage was skipped.



0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

Legnocase building

0.28g

0.07g

0.50g

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.15g

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.



Rubnerhaus building

0.28g

0.07g

0.50g

0.15g

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.



TUGraz building

0.28g

0.07g

0.50g

0.15g

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Trans.

Ref.

0.2

2

20

200

0.04 0.4 4 Spe

ctra

l ve

loci

ty [

cm/s

]

Period [s]

Long.

Ref.


Introduction - Constructive systems - Material and geometry - Tests setup - Accelerograms - Test 53

Preliminary results



Legnocase building

Sensor locations

MODE_i_j MODE_01_00 MODE_02_00 MODE_03_00 MODE_04_00 MODE_05_00 MODE_06_00

Frequency [Hz] 3.62 4.25 6.03 8.22 15.88 20.56

MODE_01_00 3.62 1.0000 0.3747 0.1033 0.1918 0.0014 0.1988

MODE_02_00 4.25 0.3747 1.0000 0.0283 0.1166 0.0191 0.0332

MODE_03_00 6.03 0.1033 0.0283 1.0000 0.1330 0.0032 0.0327

MODE_04_00 8.22 0.1918 0.1166 0.1330 1.0000 0.5394 0.0117

MODE_05_00 15.88 0.0014 0.0191 0.0032 0.5394 1.0000 0.0050

MODE_06_00 20.56 0.1988 0.0332 0.0327 0.0117 0.0050 1.0000

i Mode

j dynamic identification

Initial frequencies

Initial MAC coefficients



Mode 1: f= 3.62 Hz Mode 2: f= 4.25 Hz Mode 3: f= 6.03 Hz

Initial frequencies and mode shapes

Final frequencies and mode shapes Mode 1: f= 3.35 Hz Mode 2: f= 3.85 Hz Mode 3: f= 5.84 Hz



Rubnerhaus building

Sensor locations

Initial frequencies

Initial MAC coefficients MODE_i_j MODE_01_00 MODE_02_00 MODE_03_00 MODE_04_00 MODE_05_00 MODE_06_00

Frequency [Hz] 3.28 3.85 4.49 5.92 14.32 18.46

MODE_01_00 3.28 1.0000 0.3106 0.0785 0.0154 0.0010 0.1023

MODE_02_00 3.85 0.3106 1.0000 0.0697 0.0173 0.0032 0.0045

MODE_03_00 4.49 0.0785 0.0697 1.0000 0.1180 0.5586 0.1319

MODE_04_00 5.92 0.0154 0.0173 0.1180 1.0000 0.1380 0.0339

MODE_05_00 14.32 0.0010 0.0032 0.5586 0.1380 1.0000 0.1894

MODE_06_00 18.46 0.1023 0.0045 0.1319 0.0339 0.1894 1.0000

i Mode




TUGraz building

Sensor locations

Initial frequencies

Initial MAC coefficients MODE_i_j MODE_01_01 MODE_02_01 MODE_03_01 MODE_04_01 MODE_05_01 MODE_06_01 MODE_07_01 MODE_08_01

Frequency [Hz] 3.98 5.34 6.59 14.55 15.46 19.78 21.26 23.42

MODE_01_01 3.98 1.0000 0.0014 0.0026 0.0286 0.0428 0.0331 0.0236 0.0013

MODE_02_01 5.34 0.0014 1.0000 0.1340 0.0546 0.0010 0.0008 0.1425 0.0015

MODE_03_01 6.59 0.0026 0.1340 1.0000 0.0299 0.1405 0.0498 0.0003 0.0062

MODE_04_01 14.55 0.0286 0.0546 0.0299 1.0000 0.0307 0.0134 0.0351 0.0934

MODE_05_01 15.46 0.0428 0.0010 0.1405 0.0307 1.0000 0.0006 0.2432 0.0732

MODE_06_01 19.78 0.0331 0.0008 0.0498 0.0134 0.0006 1.0000 0.2799 0.0359

MODE_07_01 21.26 0.0236 0.1425 0.0003 0.0351 0.2432 0.2799 1.0000 0.1014

MODE_08_01 23.42 0.0013 0.0015 0.0062 0.0934 0.0732 0.0359 0.1014 1.0000

i Mode



Introduction - Constructive systems - Material and geometry - Tests setup - Accelerograms - Test

After each stage of testing a dynamic identification of the shaking table and of the model

was performed, in order to quantify the damage of the building after each seismic test.

Evaluation of the damage was also done after each test through a careful visual inspection.

60



Possible failure mechanisms of the connections between elements for TF and CLT systems

Sliding mechanisms



Overturning/rocking mechanisms



Possible failure mechanism of the walls (frame to panel connection) in the TF system

High number of small diameter steel connectors along the panel edge – high ductility and energy dissipation



The failure mechanism of the LH walls, is due to the failure of the corner joint. The structural behavior is strongly influenced by friction between the logs (pre-sliding phase)



THANKS FOR THE KIND ATTENTION!

shaking table tests at national laboratory for civil

Documents