COST ACTION TU0601...Institut Für Konstruktion und Entwurf Prof. Dr.-Ing. Ulrike Kuhlmann, Dipl.-Ing. Lars Rölle Composite joints under bending and tensionComposite joints under

Institut Für Konstruktion und Entwurf

Prof. Dr.-Ing. Ulrike Kuhlmann, Dipl.-Ing. Lars Rölle

COST ACTION TU0601Robustness of Structures

Workshop Coimbra 02./03. March 2009

Ductility of connections - a measure to improve redundancy




Redundancy within the structural system by alternate load pathsRedundancy within the structural system by alternate load paths

Alternate load path method

strengthening of one/more floors design of joints with sufficiently d til b h i d

Alternate load path method

ductile behaviour and activation of membrane forces

→ causes usually much extra costs → less extra costsd d f ll d i d→ demands for well-designed

connections2



Alternate load path methodAlternate load path method

alternate load paths may be achieved by various measures

bending capacity

plastic reserves bearing-mechanism structural behaviour

σ

actual values

σy

2D

fy,d

fy,cnominal values

actual values

σy uniaxial

ε

material hardening

tension capacity

3Dg

σy

3D

spatial

requires sufficient available ductility of members and joints3



Ductility demands to allow the activation of bearing reservesDuctility demands to allow the activation of bearing reserves

plastic material reserves plastic system reserves

large plastic strain change of bearing mechanismcatenary action

large global and local deformations

DUCTILITY

ductility of members ductility of jointsy y j

4



Type of connections influence the global resistance of the structureType of connections influence the global resistance of the structure

type of joints are usually the weakest link beam-to-column connectionin a framed structure

special focus

rigid partial-strength pinned

M M M M M M

N

MjMbeam

Φ

MjMbeam

N

Mj

Mbeam

Φ

Mj

Mbeam

N Φ

suitable but expensive suitable not suitable lack of resistanceN Φ N Φ N Φ

⇒ by adequate resistance and sufficient ductility 5



Why is sufficiently ductile joint behaviour so important?Why is sufficiently ductile joint behaviour so important?

M M

→ no redistribution possible → plastic redistribution possible

Φ Φ

6 hinge is located in the joint ⇒ high requirements concerning joint deformability



Influence of over strength effectsInfluence of over-strength effects

Over-strength effectsmaterials

b ff tmembrane effects

Adjustment of single componentsweakest components always ductile

oversizing of the bolts

avoiding premature brittle bolt failure with limited ductility of the joint7



Influence of over strength effectsInfluence of over-strength effects

Influence to the relevant component

M M Mductile component brittle component connectionM M Mductile component brittle component connection

nominal values Φ1 Φ2 Φ1+Φ2+ =Φ1 Φ2 Φ1+Φ2+ =

Φ Φ ΦcdΦ Φ Φcd

M M Mductile component brittle component connectionM M Mductile component brittle component connection

actual values + =+ =

Φ Φ ΦΦcdΦ Φ ΦΦcd

actual values of material strength may limit the deformation capacity clearly 8



Influence of over strength effects on resistance and ductility

Influence on the moment-rotation-curve

Influence of over-strength effects on resistance and ductility

IPE 500, M24 10.9, t=15 mm500

Bolf & fBolf & f

400

[kN

m] Boltmax+EPmax

Boltmax+EPmit

B l EP iΔM

Bol

Bol

Bol

fuB,(μ+σ) & fy,(μ+σ)

fuB,(μ+σ) & fy,(μ)

fuB,(μ+σ) & fy,(μ-σ)

f & f

Bol

Bol

Bol

fuB,(μ+σ) & fy,(μ+σ)

fuB,(μ+σ) & fy,(μ)

fuB,(μ+σ) & fy,(μ-σ)

f & f

200

300

nmom

ent [ Boltmax+EPmin

Boltmit+EPmax

Boltmit+EPmit

Boltmit+EPmin

ΔM(μ)

ΔM(μ+σ)

ΔM(μ-σ)

Bol

Bol

Bol

fuB,(μ) & fy,(μ+σ)

fuB,(μ) & fy,(μ)

fuB,(μ) & fy,(μ-σ)

Bol

Bol

Bol

fuB,(μ) & fy,(μ+σ)

fuB,(μ) & fy,(μ)

fuB,(μ) & fy,(μ-σ)

mom

ent

100Kno

te

Boltmit EPmin

Boltmin+EPmax

Boltmin+EPmit

Boltmin+EPmin

Bol

Bol

Bol

fuB,(μ-σ) & fy,(μ+σ)

fuB,(μ-σ) & fy,(μ)

fuB,(μ-σ) & fy,(μ-σ)

Bol

Bol

Bol

fuB,(μ-σ) & fy,(μ+σ)

fuB,(μ-σ) & fy,(μ)

fuB,(μ-σ) & fy,(μ-σ)

m

00 10 20 30 40 50 60 70

Knotenrotation [mrad]

ΔΦ(μ)ΔΦ(μ-σ) ΔΦ(μ+σ)

rotation

small influence on the resistance, clear influence on the rotation capacity9



Influence of over strength on the structural response

structural response under service conditions

Influence of over-strength on the structural response

kNm

]

Required rotation at support

oint

mom

ent [

mean value

5% - value

95% - value

depends on bending resistanceof member and joints that dependon material strength

required rotation [mrad]

jo

Variation of material strength influence also structural response10



Probabilistic analysis to assess sufficient rotation capacity

M determintion off structural steel

Probabilistic analysis to assess sufficient rotation capacity

M determintion ofM- Φ-curve

f

fyf

stochastic distribution of strength

structural steel

boltsΦavai

Φavaifu,B

strength

Φavai - Φreq = Zequation of limit state

Z < 1t ti

M

qf live loadstochastic

distribution of loads

Φreq

rotation not

sufficient

Φ

method to assess the geometrical criteria for sufficiently ductile joint behaviour11



Redundancy by allowing force redistributionRedundancy by allowing force redistribution

requires large deformations

Φ

demands for sufficient resistance (M-N), beside ductility

N N

activation of alternate load paths requires joint ductility and sufficient M-N resistance13



Experimental test simulation of steel and composite jointsExperimental test simulation of steel and composite joints

Steel Joint Tests Composite Joint Tests

bending test

M

bending and tension

M

Φ

M-Φ-curve

N

M-N-Interaction

determination of requirementsand design criteria for

highly ductile behaviour

NM

M-Φ-curvehighly ductile behaviour

14

Φ

for hogging and sagging moment



Composite joints under bending and tensionComposite joints under bending and tension

Test results

80V1 nt

M-phi-curve M-N-interaction1

20

40

60

kNm

]

V1V2V3V4V5

ggin

g m

omen

0,2

0,4

0,6

0,8

m]

-40

-20

00 100 200 300 400 500 600 700

Mom

ent [

k

g m

omen

tsa

g

0 6

-0,4

-0,2

0-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70M

[kN

m

Joint Rotation V1Joint Rotation V2

-80

-60

Normal Force [kN]

hogg

ing

-1

-0,8

-0,6

phi[mrad]

Joint Rotation V3Joint Rotation V4Joint Rotation V5

→ Joints have been able to follow the whole M-N-curve from pure bending state

to pure tension state due to sufficient ductility

15



Joint tests confirmed by a Substructure testJoint tests confirmed by a Substructure test

Φavai

[Substructure test

Liége]Liége]

structural response allows force redistribution by activation of membrane effects 16



OutlookOutlook

further numerical and experimental investigations are planned

stress

P(t)t1t

strain

P(t)t2t3

straindetermination high stress-strain rates

of plates and bolts component response under high velocity

P(t)component response under high velocity

loadingrotation capacity and strength?

including additional bearing effects3D effects

17



Thanks for attention !Thanks for attention !

Universität StuttgartInstitut für Konstruktion und EntwurfSchwerpunkte: Stahlbau, Holzbau und VerbundbauProf Dr -Ing Ulrike KuhlmannProf. Dr. Ing. Ulrike KuhlmannDipl.-Ing. Lars RöllePfaffenwaldring 770569 StuttgartGermany

Tel. +49 711 685 66245Fax +49 711 685 66236Fax +49 711 685 [email protected]

Documents

COST ACTION TU0601...Institut Für Konstruktion und Entwurf Prof. Dr.-Ing. Ulrike Kuhlmann, Dipl.-Ing. Lars Rölle Composite joints under bending and tensionComposite joints under