Lightweight Steel Framing Houses in Seismic Areas · Lightweight Steel Framing Houses in Seismic Areas Behaviour features and needs for Design Technical Regulations Dan DUBINA “Politehnica”

1

Lightweight Steel Framing Houses in Seismic Areas

Behaviour features and needs for Design Technical Regulations

Dan DUBINA “Politehnica” University of Timisoara

TURKISH CONSTRUCTIONAL STEELWORK ASSOCIATION (TUCSA)

COLD FORMED STEEL STRUCTURES WORKSHOP

Istanbul 25 March 2013

• This is a complete sustainable technology – steel

structures are 100% recyclable (It takes 6 old cars to

produce enough steel for the structure of house);

• It enables the use of high performance thermo-energetic

materials for cladding and finishing;

• It is a highly productive and qualitative technology both

for fabrication and erection;

• It enables to obtain flexible partitions;

• It enables for further up-grade, modifications and/or

development.

WHY STEEL FRAMED HOUSES?

Because:

COLD-FORMED STEEL IN RESIDENTIAL BUILDING

2

• Excellent structural performance;

• Appropriate to apply Performance Base Design and enables

for safe prediction of behavior under severe actions;

• Good robustness, easy to prevent by design the

progressive collapse;

• In case, easy to repair;

• Because their high strength/weight ratios and good

redundancy, there are the safest houses, provided it is well

designed and executed

WHY STEEL FRAMED HOUSES IN SEISMIC AREAS?

Because:


Cold-formed steel used in:

• Basement walls;

• Floor joists;

• Load bearing walls;

• Non-load bearing walls;

• Roof framing and trusses.

Construction types:

• Stick build;

• Panelized;

• Modular;

• Combination of above.

STEEL FRAMED HOUSES: CLASSICAL OR MODERN APPEARANCE


3

• Classical Appearance


• Classical Appearence


4

• Classical Appearence



• CASSETTE WALL STRUCTURE

5


• CASSETTE WALL STRUCTURE

• Wall Stud Structure


6

• Wall Stud Structure


INTRODUCTION TO SEISMIC DESIGN PRINCIPLES

DESIGN PHILOSOPHY:

• Dissipative behaviour;

• Non- or low- dissipative behaviour.

7

Design Criteria

• Ultimate limit state (collapse prevention)

• Damageability limit state (progressive design criteria in order to limit damage)

• Serviceability limit state


Dissipative structures

• Energy dissipation taking advantage of the post-elastic load bearing capacity of the structure;

• Design codes: earthquake load reduction through “q” factor.


8

Behavior factor “q” (pr EN 1998-1)

Design concept Behavior factor “q” Required

ductility class EC8

Highly dissipative structures

q4,0 H (high)

Medium dissipative structures

2,0q<4,0 M (medium)

Structures with limited

dissipation q=1,5-2,0 L (low)

Question: Thin walled steel structures are low dissipative?


Definition of the Earthquake Reduction Factor

idealized response

real response qq

qq

q1

F

Dy

d

F

F

y

Fe

Sd

De Du D

elastic response

q

R

d

S

RSds qqq

1

uRq

y

e

F

Fq

d

ys F

Fq

RSdsd qqqqqq

Ductility

Structural Overstrength

qSd-design overstrength qR -structural overstrength

Total reduction factor

Rqqq

Total reduction factor without overstrength


9

Relationships for q, q, qs

q

T (sec)

q

q

sq



• Slender sections prone to local buckling;

• Non-plastic;

• Non-dissipative.

Cold-formed steel framing for housing are usually made of class 4 or 3 sections e.g.

Seismic design codes are either restrictive or penalizing the use of these structures in seismic zones.

My

Mp

Moment

Class 1- plastic sections

Class 2- compact sections

Class 3- semi-compact sections

Class 4- slender sections

10

Earthquake behaviour of a house building

FloorDiaphragm

Steel Scheleton- member properties

- stub spacing- anchoring behaviour

Panel Dimensions- length to height

ratios

Finishing Material- material properties

- one side, both sides- tipe and position of fixings

Opening- dimensions of door

and window openings- position of openings

Boudary Conditions- corner detailes

for structural panels

Bracing- bracing typlogie- fixing detailes- pretensioning

Bracing FinishingInteraction

-added effect of bracingand exterior finishing

StructuralWall Panel

RoofDiaphragm

Ornamental elementsNon-structural Walls

Structural Behaviourof a Building


Shear Force Resisting Elements in a Wall Panel


11

An extensive research program was carried out at the “Politehnica” University of Timisoara in order to evaluate and characterize the seismic performance of cold-formed steel framed houses:

• Tests and numerical simulations on shear panels including connection behavior;

• In-situ tests of a house structure in different stages of erection.


Wall Panel Configurations

Series I and II Series III

Series IV Series OSB-I Series OSB-II

TESTING ON WALL PANELS

12

Wall Panel Typologies TESTING ON WALL PANELS

Resistance, Rigidity and Ductility of Load Bearing Wall Panels with Steel Skeleton

Series Panel

ConfigurationCladding

Testing Method

Load Vel.

(cm/min)

No. Test

O

- Monotonic 1 1

I

Corrugated Sheet LTP20/0.5 (Ext)

Monotonic 1 1 Cyclic 6 1 Cyclic 3 1

II


Gypsum Board (Int)


III

Cross Bracing (Ext-Int) Monotonic 1 1

Cyclic 3 1

IV



OSB I

10 mm OSB Panels (Ext) Monotonic 1 1

Cyclic 3 1

OSB II

10 mm OSB Panels (ext)Monotonic 1 1

Cyclic 3 1 Total Number of Specimens 15


13


Testing Procedure

• Monotonic test, with loading velocity of 1cm/min;

• From monotonic curve, determination of the Conventional Elastic Limit Displacement (Δel);

• Cyclic testing followed ECCS recommendation, (¼Δel, ½Δel, ¾Δel, 1Δel, 3x 2Δel, 3x 4Δel , 3x 6Δel,…), until significant decrease of capacity.

Ko0.4Fmax

el

0.1Ko

Fmax

2 el

4 el

el

2 el

4 el T 2T 3T 4T 5T 6T 7T 10T9T8T

el

Experimental Load-Displacement Curves

Characteristic Curves - Series III

-80000

80000

-150 150

Displacement (mm)

Fo

rce

(N

)

III-1

III-2

Characteristic Curve - Series IV

-80000

80000

-150 150

Displacement (mm)

For

ce (

N)

IV-1IV-2IV-3

Characteristic Curves - Series I

-80000

80000

-150 150

Displacement (mm)

Fo

rce

(N

)

I-1I-2I-3

Characteristic Curves - Series II

-80000

80000

-150 150

Displacement (mm)

For

ce (

N)

II-1II-2II-3

Characteristic Curves - Series OSB I

-80000

80000

-150 150

Displacement (mm)

Fo

rce

(N

)

O. I-1O. I-2

Characteristic Curves - Series OSB II

-80000

80000

-150 150

Displacement (mm)

Fo

rce

(N

)

O. II-1O. II-2


14

Failure Modes

Steel sheeting to steel framing

OSB panels to steel framing

L

D

L

D


CALIBRATION OF SIMPLIFIED HYSTERETIC MODEL

Testing: Characteristics of the Cyclic Behaviour

• Characteristics of the hysteretic behaviour:

– Nonlinearity;

– Strong pinching;

– Low Strength and Stiffness degradation;

– Performance controlled by connection.

-60000

60000

-90 90

Displacement (mm)

Fo

rce

(N

)

Envelope Curve (1st)

Stabilized Envelope (3th)

15

Testing: Equivalent models Method I - Method II

KoFel

DcurvDel DuDmax

0.1Ko

Fpl=Fu

Fmax

Fel

Ko

DuD300

DcurvDel =D400 Dmax

F300

Fpl=Fu

Fmax

• Ko up to the value of Fel (0.4Fmax);

• 0.1Ko rigidity tangent to the experimental curve;

• Fu at intersection;

• Du at intersection with the downloading branch;

• D400 corresponding to 1/400 rad panel top rotation;

• Initial stiffness (Ko), stiffness up to Fel (D400);

• Fu so as shaded areas are equal;

• Du at intersection with the downloading branch;


Testing: Initial Rigidity (Ko)

• Because of Lintel , rigidity underevaluated for specimens with openings;

0

1000

M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th

Rig

idity

- K

o (N

/mm

x m

)

Method I Method II


16

Testing: Ultimate Force (Fu)

• Strength degradation for all specimens;

• Higher values for OSB specimens;

0

10000

M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3thU

ltim

ate

For

ce -

Fu

(N/

m)

Method I Method II


Testing: Ductility (Duct= Du/Dy)

• Unexpected corner failure for specimen III-1;

• Lower ductility provided by OSB specimens;

0

5

M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th

Duc

tility

- D

uct

(mm

/mm

)

Method I Method II


17

Testing: Overstrength (Fu/Fdesign)

• Fdesign at minimum of 2/3Fu and F300 (1/300 rad); • Overstrength in the range of 1.2-1.6; • Values less relevant for panels with openings;

0

1

2

3M 1s

t

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3th M 1st

3thO

vers

tren

th -

Ful

t/F

desi

gn

(N/N

)

Method I Method II


Numerical: Proposed Model (L. Fulop, PhD Thesis, P.U. Timisoara, 2003)

2 440

3600

Fiber Hinge(FH)Hinge

M/2 M/2

• The panel is replaced by an equivalent brace system with similar hysteric behavior (DRAIN-3DX);

• Capabilities of the model:

– Pinching;

– Non-linearity;

– NO strength degradation.

Fpl

Fpl

Fel

Del

Fel

Dpl Dult


18

NONLINEAR MODEL


(University of Naples, R. Landolfo et al.)

(Richard-Abbott model parameters)

Numerical: Comparison of Experimental to FE Curve

FE Model (DRAIN) - Series I

-50000

50000

-80 -40 0 40 80

Displacement (mm)

Fo

rce

(N

)

Exp.

FEM

FE Model (DRAIN) - Series IV

-50000

50000

-140 -100 -60 -20 20 60 100 140

Displacement (mm)

Forc

e (

N)

Exp.

FEM

• Capacity limit of the Fulop’s FE model has been considered based on stabilized envelope curves;

• If few plastic are considered it can be a simple and safe approach to consider strength degradation;


19

Numerical: Interpretation of Limit States

DultDel

Fel

FyieldFmax

Loa

d

Disp.

A

B

A - envelope curveB - return path

Dyield

Yield limitElastic limit

Fmax

Fyield

Fel


Numerical: Panel Characteristics for the Analysis

Series I II IV OSB I OSB II

Wall Panel Scheme

Initial Rigidity (N/mm) 3446.6 3850.6 1766.3 4197.3 1610.5

Elastic Limit (Fel) (N)

24086 26566 128670 28942 11850

Yield Limit (Fyield) (N)

33560 39819 26812 48944 33908

Yield Limit (Dyield) (mm)

14.95 15.58 23.78 17.49 27.76

Ultimate Limit (Dult) (mm)

42.61 57.29 94.35 42.85 65.57

Duct 4.37 5.54 6.22 3.67 3.11


20

Numerical: Spectra of EQ Records for Time History Analysis

Elastic Spectra of Choosen Earthquakes

0

1

2

3

0 0.5 1 1.5 2 2.5 3

Periods (s)

Sa - S

pect

ral A

cc. (

cm/s

2) Vrancea 77

Elcentro

Newal

Shandon

Kobe

• Time history analysis using the SDOF FE models; • Masses of 2, 2.5, 3, 3.5 and 4 t; • Records scaled from 0.05g to 2g (IDA);


Numerical: Examples of Dynamic Response at Ultimate State

• Main characteristics of the hysteretic behavior are covered;

• Few plastic excursions in ultimate state (3-5);

• The approach with the stabilized envelope curves seems to be safe and reasonable;

Comp. Experimental- FEM (Series I)

-50000

0

50000

-60 -30 0 30 60

Displacement (mm)

Fo

rce

(N

)

I-3; NE(0.5g);M=3.5t

Comp. Experimental - FEM (Series II)

-60000

0

60000

-60 -30 0 30 60

Displacement (mm)

Fo

rce

(N

)

EL(0.95g);M=3.5t


21

Numerical: Series I - IDA Curves

• Important difference between earthquake level at the elastic and yield limit (overstrength);

• Less between yield and ultimate (ductility);

Kobe - Series I

0

1

2

0 1 2 3 4 5

Displacement (cm)

Sa

(g)

2 t 2.5 t3 t 3.5 t4 t

Del Dyield Dult

Elcentro - Series I

0

1

2

0 1 2 3 4 5

Displacement (cm)

Sa

(g)

2 t 2.5 t3 t 3.5 t4 t

Del Dyield Dult


Numerical: Series I -Partial ‘q’ Values

Reduction Factor Q1Series I

0

2

4

2t 2.5t 3t 3.5t 4t

Av

El

Ko

Ne

Sh

Vr

Total Reduction Factor Q1*Q2Series I

0

2

4

6

2t 2.5t 3t 3.5t 4t

Av

El

Ko

Ne

Sh

Vr

Reduction Factor Q2Series I

0

2

2t 2.5t 3t 3.5t 4t

Av

El

Ko

Ne

Sh

Vr

0

Displacement

Del Dyield Dult

Sel

Syield

Sult


22

Numerical: Summary of ‘q’ Values

Series I II IV OSB I OSB II

Scheme

q1 2.5 2.21 3.28 2.66 3.78

q2 1.46 1.65 2.36 1.38 1.88

q3 3.62 3.61 7.65 3.67 6.96

0

4

8

q1

q2

q3


Wall Panel Results

qµ = 1,4 – 1,6

qs = 2,2 – 2,6

qd = 3,0 – 4,0

The performance of the wall panels is governed by the performance of seam and sheeting to skeleton fasteners


23

Tests: Performance criteria based on fastener slip

Specimen Maximum slip in fasteners (mm)

Panel top displacement (mm)

Drift (%)

0.197 6.71 0.274 I-3

4.8 29.22 1.197 0.197 7.96 0.326

IV-2 4.8 44.13 1.808

0.197 8.11 0.332 IV-3

4.8 42.22 1.730

FO - Fully Operational: elastic deformations and small plastic elongation in fastener holes (drift<0.003);

PO - Partially Operational: local repairs of the sheeting system required against water proofing (drift<0.015);

SR - Safe but Repairs required: replacement of the sheeting is necessary, but the building is safe against collapse (drift<0.025);

PERFORMANCE OF CONNECTIONS

DEFORMATIONS OF WALL-PANELS


24

STEEL-TO-STEEL CONNECTIONS

• Representative for connection typologies used in the wall panels, two types of specimens: – corrugated sheet to skeleton (0.417 mm to 1.42 mm) using SD3-T15-4.8×22

(4.8 mm) screws (Series I-TP) – corrugated sheet to corrugated sheet (0.417 mm to 0.417 mm) using SL2-T-

A14-4.8×20 (4.8 mm) screws (Series I-TT) • Dimensions according to the ECCS P21, facilitate bearing failure of the thinner

sheet; • With the same materials and in similar conditions as the connections in the

panels; • Two loading velocities:

– V1=1mm/min for quasi static loading conditions – V2=420mm/min for high velocity tests.

Seam Connection

Sheet-to-Frame Connection


FAILURE MODES AND CHARACTERISTIC CURVES FOR “I-TP” SERIES

Characteristic curves for series I-TP-M-V1 (v=1mm/min)

0

1000

2000

3000

4000

0 5 10 15 20 25

Slip (mm)

Fo

rce

(N

)

I-TP-M-V1-1 I-TP-M-V1-2I-TP-M-V1-3 I-TP-M-V1-4I-TP-M-V1-6 I-TP-M-V1-7Aver.(v=1mm/min)

Characteristic curves for series I-TP-M-V2 (v=420mm/min)

0

1000

2000

3000

4000

0 5 10 15 20 25

Slip (mm)

Fo

rce

(N

)

I-TP-M-V2-1 I-TP-M-V2-2I-TP-M-V2-3 I-TP-M-V2-4Aver.(v=420mm/min)

Failure mode for series I-TP-V1


25

FAILURE MODES AND CHARACTERISTIC CURVES FOR “I-TT” SERIES

C ha ra c te ris t ic c urv e s fo r s e rie s I-TS -M -V1 (v =1 mm/m in)

0

5 0 0

1 0 0 0

1 5 0 0

0 5 1 0 1 5 2 0 2 5

S lip (m m )

Fo

rce

(N

)

I -T S -M -V 1 -1 I -T S -M -V 1 -2

I -T S -M -V 1 -3 I -T S -M -V 1 -4

A ve r.(v=1 m m /m in)

Characteristic curves for series I-TS-M-V2 (v=420mm/min)

0

500

1000

1500

0 5 10 15 20 25

Slip (mm)

Fo

rce

(N

)

I-TS-M-V2-1 I-TS-M-V2-2

I-TS-M-V2-3 Aver.(v=420 mm/min)

I-TT-V1

I-TT-V2


OSB-TO-STEEL CONNECTIONS

C hara cte r is t ic curves se r ies I-O P -M -V 1

0

1 0 0 0

2 0 0 0

0 2 4 6 8 1 0 1 2

S lip (m m )

Fo

rce

(N

)

I -O P -M -V 1 -1 I -O P -M -V 1 -2 I -O P -M -V 1 -3

I -O P -M -V 1 -4 I -O P -M -V 1 -5 A ve rag e

O S B -to -F ram e C o n n ec tion •Very inhomogeneous results;

•No conclusion can be drawn;

•OSB connections possess less ductility.


26

DESIGN CRITERIA FOR CONNECTIONS

Inf lue nc e o f L o ad ing V e lo c ity - S he e ting to F ram e F as te ne rs

0

2 0 0 0

4 0 0 0

0 5 1 0 1 5 2 0 2 5

S lip (m m )

Fo

rce

(N

)

E xpe rim enta l (v1=1m m /m in)

E xpe rim enta l (v2=420 m m /m in)C ha racte ris tic curve (v1 )

C ha racte ris tic curve (v2 )

De

Dr

I n f lue nc e o f L o ad ing V e lo c ity - S e am F as te ne rs

0

1 0 0 0

0 1 0 2 0

S lip (m m )

Fo

rce

(N

)

E xp e rim e nta l (v1 =1m m /m in)E xp e rim e nta l (v2 =42 0 m m /m in)C ha ra c te ris tic curve (v2 )C ha ra c te ris tic curve (v1 )

De

Dr

Characteristic curve of OSB-to-Frame Connections

0

1000

2000

0 5 10

Slip (mm)

Fo

rce

(N

)

Experimental (v1=1mm/min)Characteristic curve (v1)

De

• Significant ductility of steel-to-steel connections with the possibility to use multi level design criteria;

• Important overstrength for steel-to-steel connections;

• In case of the OSB-to-steel connections failure is non ductile without possibility to use multi level design criteria;


CONSTRUCTING THE FE MODEL

• The FE model takes into account the:

– Elastic shear deformation characteristics of the corrugated sheeting (including end distortion);

– Elastic deformation characteristics of the skeleton;

– Non-linear deformation characteristics of sheeting-to-framing and seam connections;

– Non-linear characteristics of hold-down details.


27

DEFORMATION OF THE FE MODEL


DESIGN CRITERIA TRANSLATED FROM CONNECTION TO WALL-PANEL LEVEL

• The design criteria for connections determines the criteria for panels

• Because low degradation of strength and stiffness in cyclic , monotonic analysis might apply

C o m p aris o n F E M - E xp e rim e nt (S p e c im e n I -1 )

0

7 0 0 0 0

0 0 .0 1 0 .0 2 0 .0 3

D rif t (m m /m m )

Fo

rce

(N

)

E xp . F E M

Ful

ly

Par

tially

Rep

airs

C o m p aris o n F E M - E xp e rim e nt (S p e c im e n IV -1 )

0

4 5 0 0 0

0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 0 .0 5

D rif t (m m /m m )

Fo

rce

(N

)

E xp . F E M

Ful

ly

Par

tially

Rep

airs


28

In-situ tests of a house structure in different stages of erection

IN-SITU MEASUREMENTS

Structural design based on tested panels

The skeleton with the structural OSB sheeting

Steel skeleton of the structure


29

Story Plans



Design assisted by testing

where,

Es,i = total shear force induced by seismic action on “i” direction (kN);

Rs,i = total shear wall resistance on “i” direction (kN);

Rk = characteristic strength of shear wall (kN/m);

Li = length of shear wall on “i” direction (m).

30


Design assisted by testing Perete Ax. 1

OSB pe ambele fetePerete Ax. 1'

OSB pe ambele fete

Perete Ax. 2OSB pe ambele fete

Perete Ax. 2'OSB pe ambele fete



Perete Ax. AOSB pe ambele fete

Perete Ax. BOSB pe ambele fete

Perete Ax. COSB pe ambele fete

Perete Ax. DOSB pe ambele fete

Transversal: Lgf =21.55 m, Lff =17.12 m Longitudinal: Lgf =17.22 m, Lff =12.42 m


Three subsequent erection stages

• Steel framing only;

• Steel framing with OSB panels;

• Steel framing with finishing.

Modal dynamic characteristic compared :

Numerical vs. measures

31

Modal parameters obtained by FE modeling


Mode shapes in Stage 2 (FE modeling): (a) first mode - T1, (b) second mode - T2, (c) third mode - T3

(a) (b) (c)


32

In-situ measurements (1)

Construction Stage 1 and measurement sensor location on the skeleton of the structure



Construction Stage 2 and measurement sensor location


33


Construction Stage 3 and measurement sensor location


Modal parameters based on the analysis of ambient vibrations (in different stages “S”)


34

Conclusions • Very good 2D behaviour of the wall:

• Robust , e.g. high redundancy; • Moderate ductility available ; • Performance dominated by connection

behaviour. • Structural design based on calibrated panels, e.g.

stiffness vs. capacity, confirmed • Significant contribution of finishing • Damping ratio of 5% confirmed.

Conclusions :Design Recommendations • Cold formed steel framed houses will designed as low

dissipative structures

• q= 1.5-2.0, and general rules for conceptual seismic design – EN 1998-1: regularity, continuity of axes, low stiffness and mass eccentricity , ductile connecting technology

• Elastic design of steel framing according to EN 1993-1.1, EN 1993-1-3, EN 1993-1-8

• Standard detailing

• Prescriptive method design : pre-calibrated wall-stud panel units , numerical or tests for all-steel solutions, tests for composite; calibrated models for connections allways .

• Lightweight flooring – dry or light concrete; Lightweight envelope; diaphragm capacity available enabling for box-type behaviour of the building

• Proper foundation design

Documents

Lightweight Steel Framing Houses in Seismic Areas · Lightweight Steel Framing Houses in Seismic Areas Behaviour features and needs for Design Technical Regulations Dan DUBINA “Politehnica”