Connections to EC3

1

Connections to 1993-1-8

© 2011 The Steel Construction Institute

Well, what do you expect?

Any dramatic changes? That BS 5950 was wrong?

But gravity loads reduce by ≈ 8%• Can we use BS 5950 connections?

Nominally pinned connections may also have to carry large tying forces?• How is this reconciled? – the results are not

pinned.

2


Objectives:

To reassure you about connection design

To alert you to the National Annex


EN 1993-1-8:

A bit on bolts and weld strength

A huge section on moment resistance• Like the “Green Book”

A huge section on stiffness calculation• New to the UK experience

A huge section on hollow section joints• Like CIDECT, Corus Publications etc

3


As expected:

Resistance based on the resistance of the components• Typically bolts, welds, plates

• We look for “the weakest link”


As expected:

When making assumptions about the distribution of internal forces:• The implied deformations must be realistic

• The assumed distribution of forces must acknowledge relative stiffnesses

Welds are not ductile, and bolts are not springs

4


Bolt force distribution

L1

L2

L3

2T

2T x

2T x

L2

L1

L3

L1

2T x L1

2T x x L2L2

L1

2T x x L3L3

L1

=2T( L1 + + + …… )L2²L1 L3²

L1


Bolt force distibution

5


Component strengths

All require a partial safety factor, M

… to be taken from the National Annex


From the UK National Annex:

6


From the UK National Annex:


Bolts

Plenty of grades:

4.6, 4.8, 5.6, 5.8, 6.8, 8.8, 10.9

Some limitations in the UK NA:

4.6, 4.8, 5.6, 5.8, 6.8, 8.8, 10.9

7


Bolts in shear

M2

ubvRdv, γ

AfαF

v = 0.6 for 4.6 and 8.8fub = 400 N/mm2 for 4.6

= 800 N/mm2 for 8.8A = tensile area, if threads in the shear planeA = gross area, if threads in the unthreaded shank

UK practice is fully threaded bolts, so use of the tensile area is strongly recommended


Bolts in shear

M20 8.8 bolts, A = 245 mm2

kN94.125.1

2458006.0

M2

ubvRdv,

γ

AfαF

EC3 BS 595094.1 kN 91.9 kN

8


Bolts in Shear – 8.8

21513694.160.31993-1-8

21013291.958.9BS 5950

M30M24M20M16

2 % more to the Eurocode


Bolts in bearing – to BS 5950

bspbs ptdP

So for an M 20, in 10 mm S275 plate:

kNbs 92104601020 3 P

9


Bolts in bearing – to 1993-1-8

M2

ub1Rdb, γ

tdfαkF

Effect of e2

Effect of:1. e1 and p1

2. Bearing on bolt or end plate

e2FEd

e1

p1



M2

ub1Rdb, γ

tdfαkF

kN112=10×1.25

10×20×410×81.0×11.2= 3RdF

For a M20 bolt in 10 mm S275 plate

(in a “standard” connection):

10



M2

ub1Rdb, γ

tdfαkF

For a M20 bolt in 10 mm S275 plate

(in a nearly “standard” connection):

kN164=10×1.25

10×20×410×0.1×5.2= 3RdFMaximum:


Bolts in bearing

In BS 5950, the bearing strength was arranged to limit deformation at working load to 1.5 mm• (despite it being a ULS check)

No such limit in many other codes

11


NA Note to the M table:

“in certain circumstances deformation at serviceability might control and a M2 = 1.5 would be more appropriate”

kN101.5

..23Rd 137

1020410015

FMaximum:


Bolt groups

If the shear resistance is greater than anybearing resistances, then the connection resistance = bearing

Else, the connection resistance

= n × minimum resistance

(Cl 3.7)

12


Bolt groups

Shear

Bearing (end is minimum)

Bearing (maximum)


Bolt groups

Min bearing (2) Max bearing (6)

Shear

Connection resistance

= 8 × shear

13


Bolt groups


Shear


= 8 × minimum bearing


Bolt groups


Shear


= 2 × minimum bearing + 6 × maximum bearing

14


Bolts in tension

Two approaches in BS 59501. Consider prying and use a “full” resistance

2. Ignore prying (within limits) and use a reduced resistance

198158M24

137110M20

FullReduced(8.8 bolts)


Bolts in tension

M2

sub2Rdt, γ

AfkF

k2 = 0.9

k2 = 0.63 for a countersunk bolt

kN1411.25

2458000.9Rdt,

F

137 kN in BS 5950when prying calculated

M20, 8.8

15


Bolts in tension

Two approaches in BS 59501. Consider prying and use a “full” resistance

2. Ignore prying (within limits) and use a reduced resistance

203

141

1993-1-8

198158M24

137110M20

FullReduced(8.8 bolts)


16


Bolts in tension

(8.8 bolts)

162

113

198

137

203158M24

141110M20

FullReduced at 0.8kN


Welds

A directional method

A simplified method – Cl.4.5.3.3

Fw,Ed Fw,Rd

Fw,Rd = fvw.d a (a is the throat)

fvw.d = fu is the ultimate tensile strengthof the weaker part

w depends on steel grade:= 0.85 for S 275= 0.9 for S 355

M2w

u

3

β

f

17


Welds

2

M2w

u

N/mm 222=1.25×0.85

3410

=3

γβ

f

(220 N/mm2 in BS 5950)

2

M2w

u

241N/mm=1.25×0.9

3470

=3

γβ

f

(250 N/mm2 in BS 5950)


Weld strengths

Fillet weld

(S 275)

1.54

1.155

1.2328

0.9246

TransverseLongitudinalkN/mm

+25%

+25%

18


Weld strengths

Fillet weld

(S 275)

1.25

0.94

1.538

1.156


+22%

+22%


Weld strengths

Fillet weld

(S 275)

1.25

0.94

1.54

1.155

1.531.2328

1.150.9246


19


Also in BS EN 1993-1-8

Limiting edge, end, pitch etc

Bolts through packing

Long joints

Slip-resistant connections (HSFGs)• Non-slip at SLS

• Non-slip at ULS

Block tearing (block shear)

Pins


An intermediate summary

Bolts in shear are nearly identical

Bolts in tension are nearly identical

Bolts in bearing – will not often govern

Welds are nearly identical

Plates in shear are almost identical

20


Moment-resisting connections

Eurocode term is “Rigid Joints”

The approach was captured in the “Green Book”

Software to be used!


Rigid joints

Described in the “Green Book”• Component strengths are very nearly the

same

21


Joint classification:

All connections require classification before using them:• by stiffness (elastic design)

• or by strength (plastic design)

• … or both


Joint classes

Nominally pinned

Rigid

Semi-rigid

22


From BS EN 1993-1-1

The effects of the behaviour of the joints…may generally be neglected…but where the effects are significant (such as in the case of semi-continuous joints) they shall be taken into account


Joint stiffness

Neglect in analysisAccount for in analysis

Simple= pinned

Semi-continuoususe stiffness

Continuous= rigid

Effectssignificant?

Yes No

23


Classification by Strength

Nominally pinned• Accept the rotations – must be ductile

• A capacity less than 25% of full strength

Full strength• A resistance greater than that of the

connected members


Classification by stiffness

• Nominally pinned?

• Semi-rigid?

• Rigid

Pages of calculations

24


Connection stiffness

ki is the stiffness coefficient of the component z is the lever arm is the stiffness ratio

i i

2

j 1k

EzS

j

inij,

SS


Component stiffness coefficients, k

• Web in shear

• Web in compression

• Web in tension

• Flange in bending (based on effective length of T-stub)

• End plate in bending (based on effective length of T-stub)

• Bolts in tension

25


Moment-rotation curves

2/3


Moment-rotation curves

26


Calculating stiffness

A laborious process

Some UK doubt about the results:• A BRE example from 2004

• 8mm partial depth end plate

• The connection is semi-rigid…..


Now for the good news!

The Eurocde does not insist on calculations

The UK NA is particularly helpful

27


Joint classification

A joint may be classified on the basis of experimental evidence, experience of previous satisfactory performance in similar cases or by calculations based on test evidence


UK National Annex

Connections designed in accordance with the principles in:

… the Green Book on Simple Connections –…are pinned

… the Green Book on Moment connections –…are continuous

28


Where to be concerned

Outside the UK• quels sont les livres verts?

• was sind die grüne Bücher?

• czym są zielone książki?


Where to be concerned

If you use a non-standard connection• Outside the Green Books

• With no previous satisfactory experience

• ….pinned connections carrying large tying forces?

• A golden opportunity for disputes

29


The new Green Book

Fin plate

End plate

(partial depth)

Shear (kN)

237 kN

366 kN

457 × 191 × 67, 4 rows,


The new Green Book

Fin plate

End plate

(partial depth)

Shear (kN)

237 kN

366 kN

457 × 191 × 67, 4 rows,

30


The new Green Book

Fin plate

End plate

(partial depth)

Shear (kN)

257 kN237 kN

352 kN366 kN

457 × 191 × 67, 4 rows,


The new Green Book

To overcome the tying problem with partial depth end plates

Partial depth

31


Shear and tying resistance – end plates

171

229

225

513

Old partial depth

Shear356 × 171 × 45

3 rows Tying

Tying

Shear533 × 210 × 82

5 rows

New full depth

New partial depth


The new Green Book

To overcome the tying problem

A brand new detail:• Welded to both flanges

Increases the tying force considerably

• Relatively thin platesClassified as “pinned”

Partial depth

Full depth

32



171

229

225

513

Old partial depth

220Shear356 × 171 × 45

3 rows Tying

Tying

Shear

286

384

494533 × 210 × 82

5 rows

New full depth

New partial depth



171

229

225

513

Old partial depth

220Shear356 × 171 × 45

3 rows Tying

Tying

Shear

286

384

494533 × 210 × 82

5 rows

New full depth

New partial depth

33



171

229

225

513

Old partial depth

425220Shear356 × 171 × 45

3 rows Tying

Tying

Shear

426286

602384

753494533 × 210 × 82

5 rows

New full depth

New partial depth


Bases

An effective area method

Thickness based on a cantilever around the profile

All as BS 5950

34


Splices

No change in details, but onerous “minimum” strength requirements


Splices – bearing type

Splice material should be provided to transmit 25% of the maximum compressive force in the column

35


Splices – bearing type, cover plates

Not a problem, usually

Cover plate cross section will be about 40% of capacity(much more than force)

Can be a modest problem in bolts – shear resistance reduced by packs


Splices – bearing type, cap / base

How to apply the rule, if at all.

36


Splices – non bearing type

Internal forces and moments not less that 25% of the moment capacity in both axes

2.5% on the normal force capacity


Splices – non bearing type, major axis

Cover plates generally OK

Bolts can be a problem, particularly if reduced by packs

37


Splices – non bearing type, minor axis

Cover plates generally OK• (based on cross section

resistance)

Bolt shear can often be a problem, particularly if reduced by packs

Bearing can be a problem


Splices – non bearing

More significant issues with chunky, higher grade columns

38


Joints in hollow sections

Resistance is set by choice of geometry, member etc – so a designer’s obligation to check joint strength


Joints in hollow sections

Checks based on a range of testing

Rules now found in EN 1993-1-8

39



40



41


Conclusions (1)

Components – as expected very nearly identical resistances

Connection resistances are nearly identical

So can a frame designed to EC3 have connections designed to BS 5950, if the ULS loads are given?• Yes for orthodox connections – simple or rigid


Conclusions (2)

Section classification is new in the UK• Stick with known details

Green Book “Simple” connections –summer 2011• With new, full depth end plates

Green Book “Rigid” connections – summer 2012

42


Personal view

They are inevitable

We can manage to design

SCI is the leading, independent provider of technical expertise and disseminator of best practice to the steel construction sector. We work in partnership with clients, members and industry peers to help build businesses and provide competitive advantage through the commercial application of our knowledge. We are committed to offering and promoting sustainable and environmentally responsible solutions.

Connections to EC3

Documents

steel construction instituteas

steel construction instituteen

bearing strength

uk national annex

huge section

affec3 bs

bs 5950m30m24m20m162

threaded bolts