Research Programme of the Research Fund for Coal and Steel

Steel RTD

Project carried out with a financial grant of the

Research Programme of the Research Fund for Coal and Steel

FIRE RESISTANCE OF LONG SPAN CELLULAR BEAM MADE OF

ROLLED PROFILES
Design Guide

Project Acronym: FICEB+

Grant Agreement: RFS2-CT-2007-00042

Beneficiaries: ArcelorMittal Belval & Differdange S.A., Luxembourg

ASDWESTOK Limited, United Kingdom
Steel Construction Institute, United Kingdom
CTICM, France
Université de Liège, Belgium
Ulster University, Northern Ireland

Authors: Vassart Olivier, Brasseur Marc (ArcelorMittal)

Hawes Mike (ASDWESTOK Limited)
Simms Ian (Steel Construction Institute)
Zhao Bin (CTICM)
Franssen Jean-Marc (Université de Liège)
Nadjai Ali (Ulster University)

March 2011
2 / 42
Contents
Page No
1. Analytical model for the cellular beam made of hot rolled sections in case of fire.........................5
1.1. Objectives ...............................................................................................................................5
1.2. Principles ................................................................................................................................5
1.2.1. Fire resistance for the plastic criteria .............................................................................6
1.2.1.1. Principles ...................................................................................................................6
1.2.1.2. Classification of the sections .....................................................................................8
1.2.2. Fire resistance for the instability criteria........................................................................8
1.2.2.1. Instability of a web post.............................................................................................9
1.2.2.2. Resistance to shear buckling....................................................................................10
1.2.2.3. Lateral torsional buckling ........................................................................................10
1.3. Example of application .........................................................................................................12
1.3.1. Characteristics of the beam ..........................................................................................12
1.3.2. Resistance check ..........................................................................................................13
1.3.2.1. Net section at opening no 1 : Resistance to bending moment..................................13
1.3.2.2. Net section at opening no 16 - Resistance to normal force......................................14
1.3.2.3. Net section at opening no 15 - Resistance to shear force.........................................15
1.3.2.4. Net section at opening no 12 - Interaction M-N-V ..................................................16
1.3.2.5. Shear resistance of Web post no 31 .........................................................................16
1.3.2.6. Stability of Web post no 31 .....................................................................................17
1.3.2.7. Bending resistance of gross sections........................................................................18
1.3.2.8. Shear resistance of gross sections ............................................................................18
1.3.3. Summary of the results.................................................................................................19
1.3.3.1. Checking of net sections at openings.......................................................................19
1.3.3.2. Post checking ...........................................................................................................19
1.3.3.3. Gross section checking ............................................................................................19
2. Bailey's methods extended to long span cellular beams ................................................................20
2.1. Introduction...........................................................................................................................20
2.2. Basis of design ......................................................................................................................21
2.2.1. Fire safety.....................................................................................................................21
2.2.2. Type of structure ..........................................................................................................22
2.2.3. Simple joint models......................................................................................................22
2.2.4. Floor slabs and beams ..................................................................................................23
2.2.5. Floor design zones........................................................................................................25
2.2.6. Combination of actions ................................................................................................25
2.2.7. Fire exposure ................................................................................................................27

3 / 42
 2.2.8. Fire resistance...............................................................................................................27
2.2.8.1. Natural fire (parametric temperature-time curve) ....................................................27
2.3. Recommendations for structural elements ............................................................................28
2.3.1. Floor design zones........................................................................................................28
2.3.2. Floor slab and beams....................................................................................................30
2.3.3. Fire design of floor slab ...............................................................................................30
2.3.4. Fire design of beams on the perimeter of the floor design zone. .................................33
2.3.5. Reinforcement details...................................................................................................33
2.3.6. Detailing mesh reinforcement ......................................................................................34
2.3.6.1. Detailing requirements for the edge of a composite floor slab ................................35
2.3.7. Design of non composite edge beams ..........................................................................36
2.3.8. Columns .......................................................................................................................37
2.3.9. Joints ............................................................................................................................37
2.3.9.1. Joint classification....................................................................................................38
2.3.9.2. Fire protection..........................................................................................................38
2.3.10. Overall building stability..............................................................................................38
2.4. Compartmentation.................................................................................................................38
2.4.1. Beams above fire resistant walls ..................................................................................39
2.4.2. Stability ........................................................................................................................40
2.4.3. Integrity and insulation.................................................................................................40
3. References .....................................................................................................................................41

4 / 42
1. ANALYTICAL MODEL FOR THE CELLULAR BEAM MADE OF HOT ROLLED
SECTIONS IN CASE OF FIRE

1.1. Objectives

The aim of this document is to describe the calculation methods developed to assess the resistance of
simply supported cellular beams in fire conditions. This development has been made in the scope of
the RFCS FiCEB+ [23] and in the scope of the PHD of O.Vassart [24].

This calculation procedure has been introduced in the ACB+ software available on
www.arcelormittal.com/sections

1.2. Principles

The assessment of the fire resistance of a beam consists in calculation for each of the strength criteria,
the critical temperature (for which this strength criterion is equal to 1) and the corresponding heating
up time. This calculation is made for each of the loads combinations in fire situation.

Among the strength criteria, two types are distinguished:

- the " plastic resistance " criteria, for which the resistance depends only on the steel strength
limit fy
- the " resistance to instability " criteria, for which the resistance depends on the steel strength
limit fy and on the Young modulus E.

5 / 42
Table 1-1 : Criteria taken into account for the fire resistance calculation
Criteria in Plastic Resistance
Resistance of the gross sections (at the level of the web post and filled openings):
ΓM (*): Bending resistance
ΓV: Shear resistance
ΓMV: Interaction MV
Resistance of the web posts :
ΓVh: Resistance to horizontal shear of a web post
Resistance of the net section (at the level of an opening) :
ΓM (*): Bending resistance
ΓN (*): Axial resistance
ΓV: Shear resistance
ΓMN (*): Interaction MN
ΓMV (*): Interaction MV
ΓMNV (*): Interaction MNV
(*): criteria for which a section classification is necessary.
Criteria resistance to instability
Resistance of the gross sections:
ΓVbw: Resistance to shear buckling
Resistance of the web posts :
Γb: Web post buckling resistance
Resistance of the beam :
ΓLT: Resistance to lateral torsional buckling (Only for pure steel beam).

1.2.1. Fire resistance for the plastic criteria

1.2.1.1. Principles

The principles for the calculation of the fire resistance for the plastic criteria are the following:

1. The value of the strength criterion Γ for the time 0 of the fire is calculated taking into account
the load combination chosen for the fire calculation. The calculation of the Γ is made in a
similar way than in cold conditions by replacing the partial coefficient γM0 with γM,fi see [26].

For the strength criteria dependent on the section classification, the classification differs from
the one in cold conditions (cf. 3.1.2).

2. The critical temperature associated with the value Γ obtained in 1 is calculated from the steel
strength reduction factor ky, θ given in Table 1-3.

If the value of Γ was obtained for a section of class 4, the critical temperature is calculated
with the reduction factor kp,0,2, θ, given in the following table (Table E1 of the EN 1993-1-2).

6 / 42
 Table 1-2 : Steel strength reduction factor for a class 4 section
Steel Reduction Factor
Temperature kp,0,2, θ
θ (°C)
20 1,000
100 1,000
200 0,890
300 0,780
400 0,650
500 0,530
600 0,300
700 0,130
800 0,070
900 0,050
1000 0,030
1100 0,020
1200 0,000

3. From the massivity factor associated with the considered section and from the critical
temperature calculated in 2, the heating up time of the section is calculated in a incremental
way.
4.
The following parameters are considered:

θRef: "ambient" temperature of the beam; by default θRef : = 20°C

Δt: increment of time ; by default Δt = 1 sec
ksh: correction factor for the shadow effect (value of the factor for the rebuilt section)
by default ksh = 0.7
ρa: density of the steel; ρa = 7850 kg / m3

Assuming that the temperature of the section in time ti is equal to θi, the temperature θi+1 in
time ti+1 = ti + Δt is calculated in the following way (formula (4.25) of EN 1993-1.2):

θi+1 = θi + Δθ
Am / V
Δθ = k sh h net Δt
ca ρa

Hence ca is the specific heat of the steel, calculated according to the temperature θi with the
following formulae (according to 3.4.1.2 of EN 1993-1-2 - all the relations are expressed in J /
kgK):

for 20°C ≤ θi < 600°C :

ca = 425 + 0.773 θi – 1.69 10-3 θi2 + 2.22 10-6 θi3
for 600°C ≤ θi < 735°C :
13002
c a = 666 +
738 − θ i
for 735°C ≤ θi < 900°C :
17820
c a = 545 +
θ i − 731
for 900°C ≤ θi ≤ 1200°C :

7 / 42
 ca = 650

hnet is the value of calculation of the heat flux, determined according to 3.1 of EN 1991-1-2 by
the following relations:

hnet = hnet,c + hnet,r

hnet,c is the convective part and hnet,r is the radiative part.
hnet,c = αc (θGi – θi)
hnet,c = Φ εm εf σ [(θGi+273)4 – (θi +273)4]

Where:
θGi is the hot gas temperature for the time i, calculated from the normalised ISO Curve
(Eq 3.4 of EN 1991-1-2), according to the following function:

θGi = 20 + 345 log10(8 ti + 1) [°C]

αc is the thermal transfer coefficient for convection. It’s equal by default to 25 W/m2K
(value recommended in 3.2.1 (2) of EN 1991-1-2).
Φ is the shape factor. By default equal to 1.0.
εm is the steel surface emissivity, by default equal to 0.7.
εf is the fire emissivity, by default equal to 1.0.
σ is the Boltzmann constant (= 5.67 10-8 W/m2K4)

The critical heating up time is reached when θi = θCritique.

1.2.1.2. Classification of the sections

For a criterion in fire resistance involving the classification of the studied section, the class of the
section is determined with the parameter εθ:

235
εθ = 0.85 ε = 0.85
fy

All the other parameters of the verification (in particular the reduced slenderness for the calculation of
the participating widths) remain unchanged in respect to the cold calculation.

1.2.2. Fire resistance for the instability criteria

The principles of justification of the resistance in fire condition for the instability criteria are the
following:

1. From the stresses formed by the fire load combination, the critical temperature is reached
when the instability criterion is equal to 1. The calculation of the strength criterion according
to the temperature is detailed in the following chapters. The partial safety factor γM,fi is used.
2. From the massivity criterion described below and from the critical temperature calculated in 1,
the heating up time is calculated in an incremental way according to the same method as in
2.1.

8 / 42
 The considered massivity criterions are the following ones:

- Criterion of instability of the web post: massivity of a straight web post section can be
estimated by the following value:
Am / V = 2 / tw,
where tw, is the thickness of the considered web.
- Lateral torsional buckling criterion: massivity of a “T” section at the level of an
opening for the compressed member giving the considered criterion Γ.

1.2.2.1. Instability of a web post

The criterion for resistance to buckling of an intermediate web post at elevated temperature is
given by the following equation:

σ w ,fi,Ed
Γ(θ m ) =
κ θ σ w ,fi,Rd (θ m )

It is based on the calculation of the principal stress resistance in fire situation for the half post
being studied σw,fi,Rd and the principal compressive stress in fire situation in the half post being
studied σw,fi,Ed (σw,fi,Ed,up for the upper half post and σw,fi,Ed,low for the lower half post).

σw,fi,Ed is calculated from the fire load combinations in the same way as in cold situation (see.
5.8 (5) [27]).

σw,fi,Rd, the principal stress resistance is calculated using the following formula based on
EN1993-1-2 :

χ fi ⋅ ξ ⋅ k y,θ ⋅ f y
σ w ,fi,Rd =
γ M ,fi
Where:

fy is the steel strenght limit of the considered member

γM,fi is the partial safety factor in fire condition
ξ is a shape factor for the critical section that has been calibrated using the Finite Element
modelling (see 5.8 (9) [27])
χ fi is a reduction factor for out-of-plane buckling of the web post adapted for fire situation
following EN1993-1-2, and calculated using the following formulae :
1
χ fi = 0.5
φθ + ⎛⎜ φ2θ − λ θ ⎞⎟
2
⎝ ⎠ and χ fi ≤ 1.0
2
φθ = 0.5 [1 + α λ θ + λ θ ]
235
α = 0.65
fy

The reduced non-dimensional slenderness λ θ of the web post being considered in case of fire
is given by:

9 / 42
 k y ,θ ξf yw k y ,θ
λθ = λ =
k E ,θ σ w ,fi,cr k E ,θ

Where ky,θ and kE,θ are the reduction factors for steel strength limit and Young modulus,
respectively, at elevated temperature.

λ is the non-dimensional slenderness in « cold » conditions (See 5.8 (10) [27])

The values of ky, θ and kE, θ are given in Table 1-3 (from table 3.1 of EN 1933-1-2) :

Table 1-3 : Reduction factor for the steel strenght limit and the Young Modulus
Steel temperature Reduction factor Reduction factor
θ (°C) ky,θ kE, θ
20 1,000 1,000
100 1,000 1,000
200 1,000 0,900
300 1,000 0,800
400 1,000 0,700
500 0,780 0,600
600 0,470 0,310
700 0,230 0,130
800 0,110 0,090
900 0,060 0,0675
1000 0,040 0,0450
1100 0,020 0,0225
1200 0,000 0,000

For the intermediate values of temperature, a linear interpolation is used.

In the calculation of the critical stress, the reference Euler buckling load depends on the
Young Modulus E but remains independent from the temperature.

The post-critical reserve of strength κθ is calculated from the following relation:

κθ = 1 + 0.625 (ψθ – 0.3) and 1 ≤ κθ ≤ 1.25
ψθ = ky, θ ψ

Where ψ is the non-dimensional factor calculated in the same way as in cold situation (see 5.8
(13) [27]).

1.2.2.2. Resistance to shear buckling

It is suggested not to calculate the shear buckling in fire situation.

1.2.2.3. Lateral torsional buckling

In fire situation, the composite beams are not concerned by this criterion.
As for the cold calculation, the resistance criterion for the lateral torsional buckling of the
beam in fire situation is calculated like the buckling of the compressed member. It can be
written for a member at the temperature θ:

10 / 42
 N m,fi,Ed
ΓLT (θ) =
N b,fi,Rd (θ)
Where:
Nm,fi,Ed is the normal force in the member taking into account the fire load
combination. This value is independent of the temperature θ.
Nb,fi,Ed is the resisting force to buckling of the member.
This member is the “T” shape between two lateral supports. This value depend on the
temperature θ :

Nb,fi,Rd = χfi A0 ky, θ fy / γM,fi

Where,

ky, θ is the reduction factor for the steel strength given in the Table 1-3.
A0 is the surface of the considered section at the level of the opening (“T” section) see
relation given in 5.10.1 [27].
γM,fi is the partial safety factor in fire situation
χfi is the reduction factor for lateral torsional buckling given by the following
relationships:
1
χ fi =
φ θ + (φ θ − λ θ 2 ) 0.5
2

[
φ θ = 0 .5 1 + α λ θ + λ θ 2 ]
235
α = 0.65
fy
The reduced non-dimensional slenderness λ θ considered in case of fire is given by:

k y ,θ
λθ = λ
k E ,θ

Where ky,θ and kE,θ are the reduction factors for steel strength limit and Young
modulus, respectively, at elevated temperature given in Table 1-3.
λ is the non-dimensional slenderness in « cold » conditions calculated from 5.10.1 (4)
[27]

Nota: the critical load Ncr used in the calculation of λ is independent of the
temperature and is obtained from the relationship given in 5.10.1 (5) [27].

11 / 42
1.3. Example of application

1.3.1. Characteristics of the beam

Beam: IPE400 non composite

Steel grade: S355
Span: 20m
a0: 500mm
w: 125mm
Ht: 633.8mm
Distance between beams: 3m
Permanet Load: 1kN/m2
Snow Load: 0.5kN/m2
Fire load Combinations: 1*G + 0*Q

Figure 1–1 : Geometry of the beam

12 / 42
 Figure 1–2 : Cross section of the beam

1.3.2. Resistance check

1.3.2.1. Net section at opening no 1 : Resistance to bending moment

Bending moment MEd = 11.07 kNm

Shear forces VEd,l = -34.85 kN VEd,r = -34.85 kN
Axial forces NEd,l = 0.0 kN NEd,r = 0.0 kN
Axial forces in chord Nm,sup,l = 18.17 kN Nm,sup,r = 18.17 kN
Nm,inf,l = -18.17 kN Nm,inf,r = -18.17 kN
Shear forces in chord Vm,sup,l = -17.42 kN Vm,sup,r = -17.42 kN
Vm,inf,l = -17.42 kN Vm,inf,r = -17.42 kN
Angle φ = 24.0
Partial factor γM,fi = 1.00
Yield strengths fy,top = 355 Mpa fy,bot = 355 MPa

13 / 42
 Top chord
Inclined Tee section hφ = 96.9 mm
Aφ = 3555 mm2 Avφ = 1269 mm2
Projected forces Nφ = 23.69 kN Vφ = -13.28 kN
Mφ = -2.266 kNm
Class of the chord Class 2
Bending resistant moment at 20°C Mc,Rdφ = 16.05 kNm

Criterion ΓM,fi = 0.141

Bottom chord
Inclined Tee section hφ = 96.9 mm
Aφ = 3555 mm2 Av = 1269 mm2
Projected forces Nφ = -9.515 kN Vφ = -18.56 kN
Mφ = -2.400 kNm
Class of the chord Class 1
Bending resistant moment Mc,Rdφ = 16.05 kNm
Criterion ΓM,fi = 0.150

Critical temperature ΓM : 767°C

Am/V ΓM: 150.7 m-1

1.3.2.2. Net section at opening no 16 - Resistance to normal force

Bending moment MEd = 179.7 kNm

Shear forces VEd,l = -1.124 kN VEd,r = -1.124 kN
Axial forces NEd,l = 0.0 kN NEd,r = 0.0 kN
Axial forces in chord Nm,sup,l = 295.1 kN Nm,sup,r = 295.1 kN
Nm,inf,l = -295.1 kN Nm,inf,r = -295.1 kN
Shear forces in chord Vm,sup,l = -0.562 kN Vm,sup,r = -0.562 kN
Vm,inf,l = -0.562 kN Vm,inf,r = -0.562 kN
Angle φ = 0.0
Partial factor γM,fi = 1.00
Yield strengths fy,top = 355 Mpa fy,bot = 355 MPa

Top chord
Inclined Tee section hφ = 66.9 mm
Aφ = 3078 mm2 Avφ = 990.0 mm2
Projected forces Nφ = 295.1 kN Vφ = -0.562 kN
Mφ = 0.0 kNm
Class of the chord Class 2
Normal resistant force at 20°C Nc,Rdφ = 1093 kN
Criterion ΓN,fi = 0.270

14 / 42
 Bottom chord
Inclined Tee section hφ = 66.9 mm
Aφ = 3078 mm2 Avφ = 990.0 mm2
Projected forces Nφ = -295.1 kN Vφ = -0.562 kN
Mφ = 0.0 kNm
Class of the chord Class 1
Normal resistant force at 20°C Nc,Rdφ = 1093 kN
Criterion ΓN,fi = 0.270

Critical temperature ΓN : 683°C

Am/V ΓN: 154.5 m-1

1.3.2.3. Net section at opening no 15 - Resistance to shear force

Bending moment MEd = 178.3 kNm

Shear forces VEd,l = -3.372 kN VEd,r = -3.372 kN
Axial forces NEd,l = 0.0 kN NEd,r = 0.0 kN
Axial forces in chord Nm,sup,l = 292.8 kN Nm,sup,r = 292.8 kN
Nm,inf,l = -292.8 kN Nm,inf,r = -292.8 kN
Shear forces in chord Vm,sup,l = -1.686 kN Vm,sup,r = -1.686 kN
Vm,inf,l = -1.686 kN Vm,inf,r = -1.686 kN
Angle φ= -39.0
Partial factor γM,fi = 1.00
Yield strengths fy,top = 355 MPa fy,bot = 355 MPa

Top chord
Inclined Tee section hφ = 158 mm
Aφ = 4523 mm2 Avφ = 1836 mm2
Projected forces Nφ = 226.5 kN Vφ = -76.10 kN
Mφ = 3.652 kNm
Shear resistant force at 20°C Vc,Rdφ = 376.3 kN
Criterion ΓV,fi = 0.202

Bottom chord
Inclined Tee section hφ = 158 mm
Aφ = 4523 mm2 Avφ = 1836 mm2
Projected forces Nφ = -228.6 kN Vφ = 73.48 kN
Mφ = -2.851 kNm
Shear resistant force at 20°C Vc,Rdφ = 376.3 kN
Criterion ΓV,fi = 0.195

Critical temperature ΓV : 723°C

Am/V ΓV: 145.4 m-1

15 / 42
1.3.2.4. Net section at opening no 12 - Interaction M-N-V

Bending moment MEd = 165.6 kNm

Shear forces VEd,l = -10.12 kN VEd,r = -10.12 kN
Axial forces NEd,l = 0.0 kN NEd,r = 0.0 kN
Axial forces in chord Nm,sup,l = 272.0 kN Nm,sup,r = 272.0 kN
Nm,inf,l = -272.0 kN Nm,inf,r = -272.0 kN
Shear forces in chord Vm,sup,l = -5.059 kN Vm,sup,r = -5.059 kN
Vm,inf,l = -5.059 kN Vm,inf,r = -5.059 kN
Angle φ = -21.0
Partial factor γM,fi = 1.00
Yield strengths fy,top = 355 MPa fy,bot = 355 MPa

Top chord
Inclined Tee section hφ = 89.4 mm
Aφ = 3437 mm2 Avφ = 1200.0 mm2
Projected forces Nφ = 255.8 kN Vφ = -29.31 kN
Mφ 0.163 kNm
Shear resistant force at 20°C Vc,Rdφ = 246.0 kN ΓV,fi = 0.119
Reduction ρ = 0.000 (No reduction)
Normal resistant force at 20°C NV,Rd = 1220 kN ΓNV,fi = 0.210
Bending resistant moment at 20°C MV,Rd = 14.04 kNm ΓMV,fi = 0.012
Interaction MNV ΓMNV,fi = 0.221

Bottom chord
Inclined Tee section hφ = 89.4 mm
Aφ = 3437 mm2 Avφ = 1200.0 mm2
Projected forces Nφ = -252.1 kN Vφ = -29.31 kN
Mφ = -1.335 kNm
Shear resistant force at 20°C Vc,Rdφ = 246 kN ΓV,fi = 0.158
Reduction ρ = 0.000 (No reduction)
Normal resistant force at 20°C NV,Rd = 1220 kN ΓNV = 0.249
Bending resistant moment at 20°C MV,Rd = 14.04 kNm ΓMV = 0.095
Interaction MNV ΓMNV = 0.302

Critical temperature ΓMNV : 670°C

Am/V ΓMNV: 151.5 m-1

1.3.2.5. Shear resistance of Web post no 31

Tee geometrical centres dG = 608.9 mm

Bending moments MEd,l = 32.14 kNm MEd,r = 11.07 kNm

16 / 42
 Axial forces in tees Nm,Sup,l = 52.79 kN Nm,Inf,l = -52.79 kN
Nm,Sup,r = 18.17 kN Nm,Inf,r = -18.17 kN
Horizontal shear force in post Vhm = -34.62 kN
Post width w = 125.0 mm
Resistant shear forces at 20°C VhRd,top = 220.33 kN VhRd,bot = 220.33 kN
Checkings ΓVh,top = 0.157 ΓVh,bot = 0.157

Critical temperature ΓVh : 761°C

Am/V ΓVh: 232.6 m-1

1.3.2.6. Stability of Web post no 31

Diameter a0 = 500.0 mm
Cells spacing e = 625.0 mm α = e / a0 = 1.25
Height of cross section Ht = 633.8 mm
Heights of chords hm,top = 316.9 mm hm,bot = 316.9 mm
Heights of tees hTe,top = 66.9 mm hTe,bot = 66.9 mm
Tees geometrical centres dG,top = 304.4 mm dG,bot = 304.4 mm
dG = dG,top + dG,bot dG = 608.9 mm
Area of tees A0,top = 3078.4 mm2 A0,bot = 3078.4 mm2
Shear area of tees Av0,top = 990.0 mm2 Av0,bot = 990.0 mm2
Yield strengths fy,top = 355 MPa fy,bot = 355 MPa
Shear forces VEd,l = -32.60 kN VEd,r = -34.85 kN
Moments MEd,l = 32.14 kNm MEd,r = 11.07 kNm
Shear parameters η = 0.292 kAv = 0.500
Normal forces in chords Nm,ltop = 52.79 kN Nm,lbot = -52.79 kN
Nm,rtop = 18.17 kN Nm,rbot = -18.17 kN
Shear forces in chords Vm,ltop = -16.30 kN Vm,lbot = -16.30 kN
Vm,rtop = -17.42 kN Vm,rbot = -17.42 kN
Forces in the post Vhm = -34.62 kN Mhm = 0.00 kNm
Critical section dW = 97.3 mm LW = 164.4 mm
Moments in the critical section McEd,top = -3.37 kNm McEd,bot = -3.37 kNm
Principal stresses σW,fi,top = 102 MPa σW,fi,bot = 102 MPa
Critical forces VhCr,top = 341.48 kN VhCr,bot = 341.48 kN
NmCr,top = 1533.08 kN NmCr,bot = 1533.08 kN
Critical coefficients βCr,top = 9.628 βCr,bot = 9.988
αCr,top = 9.805 αCr,bot = 9.988
Critical stresses σCr,top = 1004 MPa σCr,bot = 1023 MPa

Reduced slendernesses at 20°C λtop = 0.729 λbot = 0.723

With ξ = 1.505
Reduction factors at 20°C χtop = 0.834 χbot = 0.837
Resistant stresses at 20°C σWRd,top = 445 MPa σMPa WRd,bot = 447MPa
Plastic moments of tees at 20°C MplRd,Te,top = 8.92 kNm MplRd,Te,bot = 8.92 kNm

17 / 42
 Psi factor at 20°C Ψtop = 0.820 Ψbot = 0.820
Post-buckling factor κtop = 1.250 κbot = 1.250

Critical temperature θcrit,top = 646°C θcrit,bot = 647°C

ky,θ at critical temperature ky,θ,top = 0.3596 ky,θ,bot = 0.3572
kE,θ at critical temperature kE,θ,top = 0.229 kE,θ,bot = 0.2272
Reduced slendernesses at θcrit λθ,top = 0.92 λθ,bot = 0.92
Reduction factors at θcrit χθ,top = 0.53 χθ,bot = 0.53
Psi factor at θcrit Ψθ,top = 0.24 Ψθ,bot = 0.24

1.3.2.7. Bending resistance of gross sections

Section at web post no 16 (Section no 33)

Internal moment and force MEd = 179.86 kNm NEd = 0.00 kN
Upper flange under compression: Class 1
Class of the web
Steel fy,w = 355 MPa εw = 0.814
Slenderness: c / t = 65.67
Plastic distribution factor α = 0.50
Class of the web 2
Check of the resistance (Class2)
Steel fy,top = 355 MPa fy,bot = 355 MPa
Partial factor γM,fi = 1.00
Plastic resistant moment at 20°C Mpl,Rd = 856.24 kNm
Criterion ΓMg,fi = 0.210

Critical temperature ΓMg : 717°C

Am/V ΓMg: 185 m-1

1.3.2.8. Shear resistance of gross sections

Section at left end (Section no 1)

Height of the cross-section h = 633.8 mm
Shear area Av,top = 3140.0 mm2 Av,bot = 3140.0 mm2
Yield strengths fy,top = 355 MPa fy,bot = 355 MPa
Shear design force VEd = 35.97 kN
Shear resistance force at 20°C VplRd = 1287.14 kN γM,fi = 1.00
Criterion ΓVg = 0.028

Critical temperature ΓVg : 1060°C

Am/V ΓVg: 232.6 m-1

18 / 42
1.3.3. Summary of the results

1.3.3.1. Checking of net sections at openings

Parameter Γ Angle (°) Am/V (m-1) θcrit (°C)

ΓM 0.150 24.0 150.7 767
ΓN 0.270 0.0 154.5 683
ΓV 0.202 -39.0 145.4 723
ΓMN 0.302 21.0 151.5 670
ΓMV 0.150 24.0 150.7 767
ΓNV 0.270 0.0 154.5 683
ΓMNV 0.302 21.0 151.5 670

1.3.3.2. Post checking

Parameter Γ Am/V (m-1) θcrit (°C)

ΓVh 0.157 232.6 761
Γb - 232.6 646

1.3.3.3. Gross section checking

Parameter Γ Am/V (m-1) θcrit (°C)

ΓMg 0.210 185 717
ΓVg 0.028 232.6 1060

19 / 42
2. BAILEY'S METHODS EXTENDED TO LONG SPAN CELLULAR BEAMS

Executive summary

Large-scale fire tests conducted in a number of countries and observations of actual building fires have
shown that the fire performance of composite steel framed buildings is much better than is indicated
by fire resistance tests on isolated elements. It is clear that there are large reserves of fire resistance in
modern steel-framed buildings and that standard fire resistance tests on single unrestrained members
do not provide a satisfactory indicator of the performance of such structures.

This publication presents guidance on the application of a simple design method, as implemented in
FiCEB design spreadsheet, which has been developed as a result of observation and analysis of the
BRE Cardington large-scale building fire test program carried out during 1995 and 1996 and more
recent testing on floor slabs containing cellular beams. The recommendations are conservative and are
limited to structures similar to those tested, i.e. non-sway steel-framed buildings with composite
floors. The guidance gives designers access to whole building behaviour and allows them to
determine which members can remain unprotected while maintaining levels of safety equivalent to
traditional methods.

In recognition that many fire safety engineers are now considering natural fires, a natural fire model
may be inputted or calculated using the parametric fire method from EN1991-1-2. These options are
included alongside the use of the standard fire model; all three are expressed as temperature-time
curves.

2.1. Introduction

The design recommendations in this publication are based on the performance of composite floor
plates, observed during actual building fires and full-scale fire tests[1,2,3]. These conservative
recommendations for fire design may be considered as equivalent to advanced methods in the
Eurocodes.

Large-scale natural fire tests carried out in a number of countries have shown consistently that the
inherent fire performance of composite floor plates with unprotected steel elements is much better than
the results of standard tests with isolated elements would suggest. Evidence from real fires indicates
that the amount of protection being applied to steel elements may be excessive in some cases. In
particular, the Cardington fire tests presented an opportunity to examine the behaviour of a real
structure in fire and to assess the fire resistance of unprotected composite structures under realistic
conditions. Most test evidence is available for composite beams with plain webs but this project has
included a test on a 15m by 9m floor plate with cellular composite beams and similar good behaviour
was observed.

Where national building regulations permit performance-based design of buildings in fire, the design
method provided by this guide may be applied to demonstrate the fire resistance of the structure
without applied fire protection. In some countries acceptance of such demonstration may require
special permission from the national building control authority.

The recommendations presented in this publication can be seen as extending the fire engineering
approach in the area of structural performance and developing the concept of fire safe design. It is
intended that designs carried out in accordance with these recommendations will achieve at least the
level of safety required by national regulations while allowing some economies in construction costs.

In addition to fire resistance for the standard temperature-time curve, recommendations are presented
for buildings designed to withstand a natural fire. Natural fires can be defined using the parametric

20 / 42
temperature-time curve given in EN1991-1-2 or be user define time temperature curves from other fire
analysis software.

The recommendations apply to composite frames broadly similar to the eight-storey building tested at
Cardington, as illustrated in Figure 2–1. This project has shown that the scope may also be extended to
cellular beams fabricated from rolled sections.

Figure 2–1 : Cardington test building prior to the concreting of the floors

2.2. Basis of design

This Section gives an overview of the design principles and assumptions underlying the development
of the simple design method.

The design guidance has been developed from research based on the results from fire tests, ambient
temperature tests and finite element analyses.

2.2.1. Fire safety

The design recommendations given in the simple design method have been prepared such that the
following fundamental fire safety requirements are fulfilled:

• There should be no increased risk to life safety of occupants, fire fighters and others in the vicinity
of the building, relative to current practice.
• On the floor exposed to fire, excessive deformation should not cause failure of compartmentation,
in other words, the fire will be contained within its compartment of origin and should not spread
horizontally or vertically.

21 / 42
2.2.2. Type of structure

The design guidance given in the simple design method applies only to steel-framed buildings with
composite floor beams and slabs of the following general form:

• braced frames not sensitive to buckling in a sway mode,

• frames with connections designed using simple joint models,
• composite floor slabs comprising steel decking, a single layer of reinforcing mesh and normal or
lightweight concrete, designed in accordance with EN1994-1-1 [7],
• floor beams designed to act compositely with the floor slab and designed to EN 1994-1-1.
• cellular beams fabricated from hot rolled steel sections

The guidance does not apply to:

• floors constructed using precast concrete slabs,

• internal floor beams that have been designed to act non-compositely (beams at the edge of the
floor slab may be non-composite),
• beams with service openings (except cellular beams as defined above).

2.2.3. Simple joint models

The joint models adopted during the development of the guidance given in this publication assume
that bending moments are not transferred through the joint. The joints are known as ‘simple’.

Beam-to-column joints that may be considered as ‘simple’ include joints with the following
components:

• Flexible end plates (Figure 2–2)

• Fin plates (Figure 2–3)
• Web cleats (Figure 2–4)

Further information on the design of the components of ‘simple’ joints is given in Section 2.3.9.

Figure 2–2 : Example of a joint with flexible end plate connections

22 / 42
 Figure 2–3 : Examples of joints with fin plate connections

Figure 2–4 : Example of a joint with a web cleat connection

2.2.4. Floor slabs and beams

The design recommendations given in this guide are applicable to profiled steel decking up to 80 mm
deep with depths of concrete above the steel decking from 60 to 90 mm. The resistance of the steel
decking is ignored in the fire design method but the presence of the steel decking prevents spalling of
the concrete on the underside of the floor slab. This type of floor construction is illustrated in Figure
2–5.

The design method can be used with either isotropic or orthotropic reinforcing mesh, that is, meshes
with either the same or different areas in orthogonal directions. The steel grade for the mesh
reinforcement should be specified in accordance with EN10080. As the design method requires
ductile mesh reinforcement in order to accommodate large slab deflections Class B or Class C should
be specified. The FiCEB design spreadsheet can only be used for welded mesh reinforcement and can
not consider more than one layer of reinforcement. Reinforcement bars in the ribs of the composite
slab are not required.

The software allows user defined sizes of welded mesh the user must input the area of the mesh in
each direction. Common French and UK mesh sizes are given in the table below.

23 / 42
Table 2-1 : Fabric mesh as defined by BS 4483 [9]
Mesh Size of Weight Longitudinal wires Transverse wires
Reference mesh (kg/m2)
(mm) Size Area Size Area
(mm) (mm2/m) (mm) (mm2/m)
A142 200x200 2.22 6 142 6 142
A193 200x200 3.02 7 193 7 193
A252 200x200 3.95 8 252 8 252
A393 200x200 6.16 10 393 10 393
B196 100x200 3.05 5 196 7 193
B283 100x200 3.73 6 283 7 193
B385 100x200 4.53 7 385 7 193
B503 100x200 5.93 8 503 8 252

Table 2-2 : Fabric mesh commonly used in French market

Mesh Size of Weight Longitudinal wires Transverse wires
Reference mesh (kg/m2)
(mm) Size Area Size Area
(mm) (mm2/m) (mm) (mm2/m)
ST 20 150x300 2.487 6 189 7 128
ST 25 150x300 3.020 7 257 7 128
ST 30 100x300 3.226 6 283 7 128
ST 35 100x300 6.16 7 385 7 128
ST 50 100x300 3.05 8 503 8 168
ST 60 100x300 3.73 9 636 9 254
ST 15 C 200x200 2.22 6 142 6 142
ST 25 C 150x150 4.03 7 257 7 257
ST 40 C 100x100 6.04 7 385 7 385
ST 50 C 100x100 7.90 8 503 8 503
ST 60 C 100x100 9.98 9 636 9 636

Figure 2–5 : Cut away view of a typical composite floor construction

24 / 42
It is important to define the beam sizes used in the construction of the cellular beams within the floor
plate as this will influence the fire performance of the floor plate. The designer will need to have
details of the serial size, steel grade and degree of shear connection available for the top and bottom
tee of the internal cellular beams. The FiCEB spreadsheet allows the user to choose from a predefined
list of serial sizes covering common British and European I and H sections.

2.2.5. Floor design zones

The design method requires the designer to split the floor plate into a number of floor design zones as
shown in Figure 2–6. The beams on the perimeter of these floor design zones must be designed to
achieve the fire resistance required for the floor plate and will therefore normally be fire protected.

A floor design zone should meet the following criteria:

• Each zone should be rectangular.

• Each zone should be bounded on all sides by beams.
• The beams within a zone should only span in one direction.
• Columns should not be located within a floor design zone; they may be located on the perimeter of
the floor design zone.
• For fire resistance periods in excess of 60 minutes, or when using the parametric temperature-time
curve, all columns should be restrained by at least one fire protected beam in each orthogonal
direction.

All internal beams within the zone may be left unprotected, provided that the fire resistance of the
floor design zone is shown to be adequate using the FiCEB spreadsheet. The size and spacing of these
unprotected beams are not critical to the structural performance in fire conditions.

An example of a single floor design zone is given in Figure 2–6.

Unprotected
beam

Fire protected
beam

Figure 2–6 : Example of a floor design zone

2.2.6. Combination of actions

The combination of actions for accidental design situations given in 6.4.3.3 and Table A1.3 of
EN 1990 [11] should be used for fire limit state verifications. With only unfavourable permanent
actions and no prestressing actions present, the combination of actions to consider is:

∑G k, j ,sup + Ad + (ψ 1,1 or ψ 2,1 )Qk,1 + ∑ψ 2 ,i Q k,i

25 / 42
Where:
Gk,j,sup Unfavourable permanent action
Ad Accidental action
Qk,1 and Qk,i Accompanying variable actions, main and other respectively
ψ 1,1 Factor for the frequent value of the leading variable action
ψ 2,i Factor for the quasi-permanent value of the ith variable action

The use of either ψ1,1 or ψ2,1 with Qk,1 should be specified in the relevant National Annex. The
National Annex for the country where the building is to be constructed should be consulted to
determine which factor to use.

The values used for the ψ factors relate to the category of the variable action they are applied to. The
Eurocode recommended values for the ψ factors for buildings are given in Table A1.1 of EN 1990;
those values are confirmed or modified by the relevant National Annex. The ψ factor values for
buildings in the UK and France are summarised in Table 2-3. For floors that allow loads to be
laterally distributed, the following uniformly distributed loads are given for moveable partitions in
6.3.1.2(8) of EN 1991-1-1 [12]:

• Movable partitions with a self-weight ≤ 1,0 kN/m wall length: qk = 0,5 kN/m2
• Movable partitions with a self-weight ≤ 2,0 kN/m wall length: qk = 0,8 kN/m2
• Movable partitions with a self-weight ≤ 3,0 kN/m wall length: qk = 1,2 kN/m2.
• Moveable partitions with self-weights greater than 3.0 kN/m length should be allowed for by
considering their location.

The Eurocode recommended values for variable imposed loads on floors are given in Table 6.2 of
EN 1991-1-1; those values may also be modified by the relevant National Annex. Table 2-4 presents
the Eurocode recommended values and the values given in the UK and French National Annexes for
the imposed load on an office floor.

Table 2-3 : Values of ψ factors

Actions Eurocode recommended UK National French National
values Annex values Annex values

ψ1 ψ2 ψ1 ψ2 ψ1 ψ2

Domestic, office and 0.5 0.3 0.5 0.3 0.5 0.3

traffic areas where:
30 kN < vehicle
weight ≤ 160 kN
Storage areas 0.9 0.8 0.9 0.8 0.9 0.8
Other* 0.7 0.6 0.7 0.6 0.7 0.6

* Climatic actions are not included

26 / 42
Table 2-4 : Imposed load on an office floor
Category of Eurocode recommended UK National Annex French National Annex
loaded area values values values
qk (kN/m2) Qk (kN) qk (kN/m2) Qk (kN) qk (kN/m2) Qk (kN)
B – Office 3.0 4.5 2.5* or 2.7 3.5 – 5.0 15.0
areas 3.0**

* Above ground floor level

**At or below ground floor level

2.2.7. Fire exposure

The recommendations given in the simple design method may be applied to buildings in which the
structural elements are considered to be exposed to a standard temperature-time curve or parametric
temperature-time curve, both as defined in EN 1991-1-2. Advanced model may also be used to define
a temperature –time curve for a natural fire scenario. The resulting temperature-time time curve may
be input to the ‘User defined’ worksheet on the FiCEB spreadsheet.

In all cases, the normal provisions of national regulations regarding means of escape should be
followed.

2.2.8. Fire resistance

The Cardington fire tests were conducted using both real (‘natural’) fires and non standard gas fires.
The tests did not follow the standard temperature-time curve that is used to define the fire resistance
periods given in national regulations. Design temperatures in terms of the standard fire resistance
temperature-time curve must therefore be calculated using thermal analysis.

The recommended periods of fire resistance for elements of construction in various types of building
may be found in national regulations. The structural elements of most two-storey buildings require 30
minutes fire resistance and those in most buildings between three and five storeys require 60 minutes
fire resistance.

The following recommendations may be applied to buildings in which the elements of structure are
required to have up to 120 minutes fire resistance. Provided that they are followed, composite steel
framed buildings will maintain their stability for this period of fire resistance, when any compartment
is subject to the standard temperature-time curve [1].

All composite steel framed buildings with composite floors may be considered to achieve 15 minutes
fire resistance without fire protection, and so no specific recommendations are given in this case.

2.2.8.1. Natural fire (parametric temperature-time curve)

The FiCEB software allows the effect of natural fire on the floor plate to be considered using the
parametric temperature-time curve as defined in EN1991-1-2 Annex A [25]. It should be noted that
this is an Informative Annex and its use may not be permitted in some European countries, such as
France. Before final design is undertaken the designer should consult the relevant National Annex.

27 / 42
Using this parametric fire curve, the software defines the compartment temperature taking account of:
• The compartment size:
• Compartment length
• Compartment width
• Compartment height

The height and area of windows:

• Window height
• Window length
• Percentage open window

The amount of combustibles and their distribution in the compartment

• Fire Load
• Combustion factor
• The rate of burning
• The thermal properties of the compartment linings

The temperature of a parametric fire will often rise more quickly than the standard fire in the early
stages but, as the combustibles are consumed, the temperature will decrease rapidly. The standard fire
steadily increases in temperature indefinitely.

The standard temperature-time curve and a typical parametric temperature-time curve are shown in
Figure 2–7.

1200
Parametric

1000
Temperature [ o C ]

800

Standard
600

400

200

0
0 15 30 45 60 75 90
Time [mins]

Figure 2–7 : Comparison of typical parametric and standard temperature-time curve

2.3. Recommendations for structural elements

2.3.1. Floor design zones

Each floor should be divided into design zones that meet the criteria given in Section 2.2.5.

28 / 42
The division of a floor into floor design zones is illustrated in Figure 2–8. Floor zones designated ‘A’
are within the scope of the design guide and their load bearing performance in fire conditions may be
determined using the FiCEB spreadsheet. The zone designated ‘B’ is outside the scope of the software
because it contains a column and the beams within the zone do not all span in the same direction.

A single floor zone is illustrated in Figure 2–9 showing the beam span designations used in the FiCEB
software. Normal design assumes that floor loads are supported by secondary beams which are
themselves supported on primary beams.

The fire design method assumes that at the fire limit state, the resistance of the unprotected internal
beams reduces significantly, leaving the composite slab as a two way spanning element simply
supported around its perimeter. In order to ensure that the slab can develop membrane action, the
FiCEB spreadsheet computes the moment applied to each perimeter beam as a result of the actions on
the floor design zone. To maintain the vertical support to the perimeter of the floor design zone in
practice, the degree of utilisation and hence the critical temperature of these perimeter beams must be
calculated using appropriate cellular beam design software. The fire protection for these beams should
be designed on the basis of this critical temperature and the fire resistance period required for the floor
plate in accordance with national regulations.

As noted in Section 2.2.4, a restriction on the use of the FiCEB spreadsheet is that for 60 minutes or
more fire resistance, the zone boundaries should align with the column grid and the boundary beams
should be fire protected. For 30 minutes fire resistance, this restriction does not apply and the zone
boundaries do not have to align with the column grid. For example, in Figure 2–8, zones A2 and A3
have columns at only two of their corners and could only be considered as design zones for a floor that
requires no more than 30 minutes fire resistance.

A(2)
Stairs

Stairs Core

A(3) A(1)

Key to figure
A: These zones may be designed using FiCEB B: Outside the scope of FiCEB
A(1) Any period of fire resistance
A(2) & A(3) only 30 minutes fire resistance

Figure 2–8 : Possible floor design zones

29 / 42
 L1
SIDE A

Unprotected
internal
SIDE D beams

SIDE B
L2

Protected
perimeter
beams

SIDE C

Figure 2–9 : Definition of span 1 (L1) and span 2 (L2) and the beam layout for a floor design
zone in a building requiring fire resistance of 60 minutes or more.

2.3.2. Floor slab and beams

The FiCEB spreadsheet calculates the load bearing capacity of the floor slab and unprotected beams at
the fire limit state. The simple design method, implemented in the software assumes that each floor
design zone will have adequate support on its perimeter. This is achieved in practice by fire protecting
the beams on the perimeter of each floor design zone. To ensure that adequate fire protection is
provided, the software calculates the critical temperature for each perimeter beam based on the loading
applied to the floor design zone.

2.3.3. Fire design of floor slab

Load bearing performance of the composite floor slab

When calculating the load bearing capacity of each floor design zone the resistance of the composite
slab and the unprotected cellular beams are calculated separately. The slab is assumed to have no
continuity along the perimeter of the floor design zone. The load that can be supported by the flexural
behaviour of the composite slab within the floor design zone is calculated based on a lower bound
mechanism assuming a yield line pattern as shown in Figure 2–10.
Yield lines

Simply supported
on 4 edges

Figure 2–10 : Assumed yield line pattern used to calculate slab resistance

30 / 42
The value of the resistance calculated using the lower bound mechanism is enhanced by considering
the beneficial effect of tensile membrane action at large displacements. This enhancement increases
with increasing vertical deflection of the slab until failure occurs due to fracture of the reinforcement
across the short slab span or compressive failure of the concrete in the corners of the slab, as shown by
Figure 2–11. As the design method can not predict the point of failure, the value of deflection
considered when calculating the enhancement is based on a conservative estimate of slab deflection
that includes allowance for the thermal curvature of the slab and the strain in the reinforcement, as
shown below.

α (T2 − T1 )l 2 ⎛ 0 .5 f y ⎞ 3L2
w= + ⎜⎜ ⎟
⎟ 8
19.2h ⎝ Ea ⎠

The deflection allowed due to elongation of the reinforcement is also limited by the following
expression.

α (T2 − T1 )l 2 l
w≤ +
19.2h 30

Where
(T2 – T1) is the temperature difference between the top and bottom surface of the slab
L is the longer dimension of the floor design zone
l is the shorter dimension of the floor design zone
fy is the yield strength of the mesh reinforcement
E is the modulus of elasticity of the steel
h is the overall depth of the composite slab
α is the coefficient of thermal expansion of concrete

All of the available test evidence shows that this value of deflection will be exceeded before load
bearing failure of the slab occurs. This implies that the resistance predicted using the design method
will be conservative compared to its actual performance.

The overall deflection of the slab is also limited by the following expression.

L+l
w≤
30

31 / 42
 Full depth crack Compression failure of concrete

Reinforcement in
longer span fractures

Yield-line pattern Edge of slab moves towards centre

of slab and 'relieves' the strains in
the reinforcement in the short span

(a) Tensile failure of the reinforcement

Concrete crushing due

to in-plane stresses

Yield-line pattern Edge of slab moves towards centre

of slab and 'relieves' the strains in
the reinforcement in the short span
(b) Compressive failure of the concrete

Figure 2–11 : Failure mode due to fracture of the reinforcement

The residual bending resistance of the unprotected composite beams is then added to the enhanced
slab resistance to give the total resistance of the complete system.

Integrity and insulation performance of the composite slab

The FiCEB spreadsheet does not explicitly check the insulation or integrity performance of the floor
slab. The designer must therefore ensure that the slab thickness chosen is sufficient to provide the
necessary insulation performance in accordance with the recommendations given in EN 1994-1-2.

To ensure that the composite slab maintains its integrity during the fire and that membrane action can
develop, care must be taken to ensure that the reinforcing mesh is properly lapped. This is especially
important in the region of unprotected beams and around columns. Further information on required
lap lengths and placement of the reinforcing mesh is given in Section 2.3.5.

32 / 42
2.3.4. Fire design of beams on the perimeter of the floor design zone.

The beams along the perimeter of the floor design zone, labelled A to D in Figure 2–9, should achieve
the fire resistance required for the floor plate, in order to provide the required vertical support to the
perimeter of the floor design zone. This usually results in these beams being fire protected.

The FiCEB spreadsheet calculates the design effect of actions on these perimeter beams and reports
this in the output. In order to determine the required fire protection of these beams the room
temperature moment of resistance of the beam must be calculated, in order to calculate the degree of
utilisation for each perimeter beam, which is calculated using the guidance given in EN 1993-1-2
§4.2.4, as shown below.
E
μ 0 = fi,d
Rfi,d,0
Where
Efi,d is the design effect of actions on the beam in fire
Rfi,d,0 is the design resistance of the beam at time t = 0

Having calculated the degree of utilisation, the critical temperature of the bottom flange of the
perimeter beams may be calculated using cellular beam design software. This critical temperature
should be use when specifying the fire protection required by each of the perimeter beams on the floor
design zone.

When specifying fire protection for the perimeter beams, the fire protection supplier must be given the
section factor for the member to be protected and the period of fire resistance required and the critical
temperature of the member. Most reputable fire protection manufacturers will have a multi
temperature assessment for their product which will have been assessed in accordance with EN 13381-
4[13] for non-reactive materials or EN 13381-8[14] for reactive materials (intumescents). Design
tables for fire protection which relate section factor to protection thickness are based on a single value
of assessment temperature. This assessment temperature should be less than or equal to the critical
temperature of the member.

2.3.5. Reinforcement details

The yield strength and ductility of the reinforcing steel material should be specified in accordance with
the requirements of EN 10080. The characteristic yield strength of reinforcement to EN 10080 will be
between 400 MPa and 600 MPa, depending on the national market. In order that the reinforcement
has sufficient ductility to allow the development of tensile membrane action, Class B or Class C
should be specified.

In most countries, national standards for the specification of reinforcement may still exist as non-
contradictory complimentary information (NCCI), as a common range of steel grades have not been
agreed for EN 10080.

In composite slabs, the primary function of the mesh reinforcement is to control the cracking of the
concrete. Therefore the mesh reinforcement tends to be located as close as possible to the surface of
the concrete while maintaining the minimum depth of concrete cover required to provide adequate
durability, in accordance with EN 1992-1-1[0]. In fire conditions, the position of the mesh will affect
the mesh temperature and the lever arm when calculating the bending resistance. Typically, adequate
fire performance is achieved with the mesh placed between 15 mm and 45 mm below the top surface
of the concrete.

33 / 42
Section 2.3.6 gives general information regarding reinforcement details. Further guidance and
information can be obtained from, EN 1994-1-1 [7] and EN 1994-1-2[6] or any national specifications
such as those given in reference [20].

2.3.6. Detailing mesh reinforcement

Typically, sheets of mesh reinforcement are 4.8 m by 2.4 m and therefore must be lapped to achieve
continuity of the reinforcement. Sufficient lap lengths must therefore be specified and adequate site
control must be put in place to ensure that such details are implemented on site. Recommended lap
lengths are given in section 8.7.5 of EN1992-1-1[19] or can be in accordance with Figure 2–8. The
minimum lap length for mesh reinforcement should be 250 mm. Ideally, mesh should be specified
with ‘flying ends’, as shown in Figure 2–12, to eliminate build up of bars at laps. It will often be
economic to order ‘ready fit fabric’, to reduce wastage.

Flying
ends

Figure 2–12 : Mesh with flying ends

Table 2-5 : Recommended tension laps and anchorage lengths for welded mesh
Reinforcement Type Wire/Bar Type Concrete Grade
LC NC LC NC LC NC
25/28 25/30 28/31 28/35 32/35 32/40
Grade 500 Bar of Ribbed 50d 40d 47d 38d 44d 35d
diameter d
6 mm wires
Ribbed 300 250 300 250 275 250
7 mm wires
Ribbed 350 300 350 275 325 250
8 mm wires
Ribbed 400 325 400 325 350 300
10 mm wires
Ribbed 500 400 475 400 450 350
Notes:
These recommendations can be conservatively applied to design in accordance with EN 1992-1-1.
Where a lap occurs at the top of a section and the minimum cover is less than twice the size of the lapped reinforcement, the
lap length should be increased by a factor of 1.4.
Ribbed Bars/Wires are defined in EN 10080
The minimum Lap/Anchorage length for bars and fabric should be 300 mm and 250 mm respectively.

34 / 42
2.3.6.1. Detailing requirements for the edge of a composite floor slab

The detailing of reinforcement at the edge of the composite floor slab will have a significant effect on
the performance of the edge beams and the floor slab in fire conditions. The following guidance is
based on the best practice recommendations for the design and construction of composite floor slabs to
meet the requirements for room temperature design. The fire design method and guidance presented in
this document assumes that the composite floor is constructed in accordance with these
recommendations.
Edge trim should be set out from
centre line of beam (not grid)
C
L Beam
Decking

Figure 2–13 : Setting out of edge trim

The edge of the composite slab is usually formed using ‘edge trims’ made from strips of light gauge
galvanized steel fixed to the beam in the same way as the decking, as shown in Figure 2–13. In cases
where the edge beam is designed to act compositely with the concrete slab, U shaped reinforcing bars
are required to prevent longitudinal splitting of the concrete slab. These reinforcement bars also
ensure that the edge beam is adequately anchored to the slab when using this simple design method.

Some typical slab edge details covering the two deck orientations are given in Figure 2–14. Where the
decking ribs run transversely over the edge beam and cantilevers out a short distance, the edge trim
can be fastened in the manner suggested in Figure 2–14(a). The cantilever projection should be no
more than 600 mm, depending on the depth of the slab and deck type used.

The more difficult case is where the decking ribs run parallel to the edge beam, and the finished slab is
required to project a short distance, so making the longitudinal edge of the sheet unsupported Figure
2–14(b). When the slab projection is more than approximately 200 mm (depending on the specific
details), the edge trim should span between stub beams attached to the edge beam, as shown in Figure
2–14(c). These stub beams are usually less than 3 m apart, and should be designed and specified by
the structural designer as part of the steelwork package.’

35 / 42
 Mesh reinforcement Restraint strats at
600 mm c/c approx.

Minimum 114 mm
(for 19 mm studs)
Additional U-bars required to
resist longitudinal splitting 75 mm

Maximum 600 mm
cantilever (or 1/4 of
adjacent span, if less)

a) Typical end cantilever

(decking ribs transverse to beam)

Additional U-bars required to

U-bars required to prevent resist longitudinal splitting
longitudinal splitting
Fixing to top Restraint straps at
of edge trim 600 mm c/c approx.

Restraint straps at
Fixing 600 mm c/c approx.
Stub cantilever
Max. 200 mm specified by Steel deck cut on site
structural designer to suit edge detail

> 200 mm

b) Typical edge detail c) Side cantilever with stub bracket

(decking ribs parallel to beam) (decking ribs parallel to beam)

Figure 2–14 : Typical edge details

2.3.7. Design of non composite edge beams

It is common practice for beams at the edge of floor slabs to be designed as non composite beams.
This is because the costs of meeting the requirements for transverse shear reinforcement are more than
the costs of installing a slightly heavier non composite beam. For fire design, it is important that the
floor slab is adequately anchored to the edge beams, as these beams will be at the edge of floor design
zones. Although not usually required for room temperature design of non composite edge beams, this
guide recommends that shear connectors are provided at not more than 300 mm centres and U shaped
reinforcing bars positioned around the shear connectors, as described in Section 2.3.6.1.

Edge beams often serve the dual function of supporting both the floors and the cladding. It is
important that the deformation of edge beams should not affect the stability of cladding as it might
increase the danger to fire fighters and others in the vicinity. (This does not refer to the hazard from
falling glass that results from thermal shock, which can only be addressed by use of special materials
or sprinklers.) Excessive deformation of the façade could increase the hazard, particularly when a
building is tall and clad in masonry, by causing bricks to be dislodged.

36 / 42
2.3.8. Columns

The design guidance in this document is devised to confine structural damage and fire spread to the
fire compartment itself. In order to achieve this, columns (other than those in the top storey) should be
designed for the required period of fire resistance or designed to withstand the selected natural
(parametric) fire.

Any applied fire protection should extend over the full height of the column, including the connection
zone (see Figure 2–15). This will ensure that no local squashing of the column occurs and that
structural damage is confined to one floor.

Bolt cleats
do not require
Protection to
protection
underside of
floor slab

Figure 2–15 : Extent of fire protection to columns

In the Cardington fire tests, the protected columns performed well with no sign of collapse. However,
subsequent finite element modelling has indicated the possibility that premature column failure could
occur in some circumstances. A mode of behaviour has been identified (0) in which expansion of the
floors induces moments in the columns. This can have the effect of reducing the temperature at which
a column would fail.

It is recommended that, as a conservative measure, the protection to the columns at the edge of the
floor plate in buildings of more than two storeys should be increased by basing its thickness on a
critical temperature of 500°C, or 80ºC less than the critical temperature given in EN 1993-1-2,
whichever is the lower.

For most board fire protection materials, this reduction in critical temperature will have no effect, as
the minimum available thickness of board will suffice.

2.3.9. Joints

As stated in Section 2.2.3 the values given by the design method relate to ‘simple’ joints such as those
with flexible end plates, fin plates and web cleats.

The steel frame building tested at Cardington contained flexible end plate and fin plate connections.
Partial and full failures of some of the joints were observed during the cooling phase of the Cardington
fire tests; however, no failure of the structure occurred as a result.

In the case where the plate was torn off the end of the beam, no collapse occurred because the floor
slab transferred the shear to other load paths. This highlights the important role of the composite floor
slab, which can be achieved with proper lapping of the reinforcement.

The resistances of the simple joints should be verified using the rules given in EN 1993-1-8[19].

37 / 42
2.3.9.1. Joint classification

Joint details should be such that they fulfil the assumptions made in the design model. Three joint
classifications are given in EN 1993-1-8:
• Nominally pinned
− Joints that transfer internal shear forces without transferring significant moments.
• Semi-rigid
− Joints that do not satisfy the nominally pinned nor the rigid joint criteria.
• Rigid
− Joints that provide full continuity.

EN 1993-1-8 §5.2 gives principles for the classification of joints based on their stiffness and strength;
the rotation capacity (ductility) of the joint should also be considered.

As stated in Section 2.2.3 the values given by the simple design method have been prepared assuming
the use of nominally pinned (simple) joints. To ensure that a joint does not transfer significant
bending moments and so that it is a ‘simple’ joint it must have sufficient ductility to allow a degree of
rotation. This can be achieved by detailing the joint such that it meets geometrical limits.

2.3.9.2. Fire protection

In cases where both structural elements to be connected are fire protected, the protection appropriate to
each element should be applied to the parts of the plates or angles in contact with that element. If only
one element requires fire protection, the plates or angles in contact with the unprotected elements may
be left unprotected.

2.3.10. Overall building stability

In order to avoid sway collapse, the building should be braced by shear walls or other bracing systems.
Masonry or reinforced concrete shear walls should be constructed with the appropriate fire resistance.

If bracing plays a major part in maintaining the overall stability of the building it should be protected
to the appropriate standard.

In two-storey buildings, it may be possible to ensure overall stability without requiring fire resistance
for all parts of the bracing system. In taller buildings, all parts of the bracing system should be
appropriately fire protected.

One way in which fire resistance can be achieved without applied protection is to locate the bracing
system in a protected shaft such as a stairwell, lift shaft or service core. It is important that the walls
enclosing such shafts have adequate fire resistance to prevent the spread of any fire. Steel beams,
columns and bracing totally contained within the shaft may be unprotected. Other steelwork
supporting the walls of such shafts should have the appropriate fire resistance.

2.4. Compartmentation

National regulations require that compartment walls separating one fire compartment from another
shall have stability, integrity and insulation for the required fire resistance period.

Stability is the ability of a wall not to collapse. For loadbearing walls, the loadbearing capacity must
be maintained.

Integrity is the ability to resist the penetration of flames and hot gases.

38 / 42
Insulation is the ability to resist excessive transfer of heat from the side exposed to fire to the
unexposed side.

2.4.1. Beams above fire resistant walls

When a beam is part of a fire resisting wall, the combined wall/beam separating element must have
adequate insulation and integrity as well as stability. For optimum fire performance, compartment
walls should, whenever possible, be located beneath and in line with beams.

Beams in the wall plane

The Cardington tests demonstrated that unprotected beams above and in the same plane as separating
walls (see Figure 2–16), which are heated from one side only, do not deflect to a degree that would
compromise compartment integrity, and normal movement allowances are sufficient. Insulation
requirements must be fulfilled and protection for 30 or 60 minutes will be necessary; all voids and
service penetrations must be fire stopped. Beams protected with intumescent coatings require
additional insulation because the temperature on the non fire side is likely to exceed the limits required
in the fire resistance testing standards[21,22].

Protection to
beam (spray
or board)
Normal
deflection
head

Compartment wall

Figure 2–16 : Beams above and in line with walls

Beams through walls

The Cardington tests showed that floor stability can be maintained even when unprotected beams
suffer large deflections. However, when walls are located off the column grid, large deflections of
unprotected beams can compromise integrity by displacing or cracking the walls through which they
pass. In such cases, the beams should either be protected or sufficient movement allowance provided.
It is recommended that a deflection allowance of span/30 should be provided in walls crossing the
middle half of an unprotected beam. For walls crossing the end quarters of the beam, this allowance
may be reduced linearly to zero at end supports (see Figure 2–17). The compartment wall should
extend to the underside of the floor.

39 / 42
 Deformable detail

Compartment w all

Figure 2–17 : Deformation of beams crossing walls

2.4.2. Stability

Walls that divide a storey into more than one fire compartment must be designed to accommodate
expected structural movements without collapse (stability). Where beams span above and in the plane
of the wall, movements, even of unprotected beams, may be small and the normal allowance for
deflection should be adequate. If a wall is not located at a beam position, the floor deflection that the
wall will be required to accommodate may be large. It is therefore recommended that fire
compartment walls should be located at a beam positions whenever possible.

In some cases, the deflection allowance may be in the form of a sliding joint. In other cases, the
potential deflection may be too large and some form of deformable blanket or curtain may be required,
as illustrated in Figure 2–17.

National recommendations should be consulted for the structural deformations which should be
considered when ensuring that compartmentation is maintained.

2.4.3. Integrity and insulation

Steel beams above fire compartment walls are part of the wall and are required to have the same
separating characteristics as the wall. A steel beam without penetrations will have integrity.
However, any service penetrations must be properly fire stopped and all voids above composite beams
should also be fire stopped.

40 / 42
3. References

[1] BAILEY, C. G. and MOORE, D. B.

The structural behaviour of steel frames with composite floor slabs subject to fire, Part 1:
Theory
The Structural Engineer, June 2000
[2] BAILEY, C. G. and MOORE, D. B.
The structural behaviour of steel frames with composite floor slabs subject to fire, Part 2:
Design
The Structural Engineer, June 2000
[3] BAILEY, C. G
Membrane action of slab/beam composite floor systems in fire
Engineering Structures 26
[4] EN 1991-1-2:2002 Eurocode 1: Actions on structures – Part 1 2: General actions. Actions on
structures exposed to fire
CEN
[5] EN 1993-1-2:2005 Eurocode 3. Design of steel structures. General rules. Structural fire design
CEN
[6] EN 1994-1-2:2005 Eurocode 4. Design of composite steel and concrete structures. Structural
fire design
CEN
[7] EN 1994-1-1:2004 Eurocode 4: Design of composite steel and concrete structures – Part 1 1:
General rules and rules for buildings
CEN
[8] EN 10080:2005 Steel for the reinforcement of concrete - Weldable reinforcing steel – General,
CEN.
[9] BS 4483:2005 Steel fabric for the reinforcement of concrete. Specification. BSI
[10] BS 4449:1:2005 Steel for the reinforcement of concrete. Weldable reinforcing steel. Bar, coil
and decoiled product. Specification
BSI
[11] EN 1990:2002 Eurocode – Basis of structural design
CEN
[12] EN 1991-1-1:2003 Eurocode 1: Actions on structures – Part 1-1: General actions – Densities,
self-weight, imposed loads for buildings
CEN
[13] EN13381-4 Test methods for determining the contribution to the fire resistance of structural
members. Applied passive protection to steel members, CEN, (To be published 2009)
[14] EN13381-8 Test methods for determining the contribution to the fire resistance of structural
members. Applied reactive protection to steel members, CEN, (To be published 2009)
[15] EN 1992-1-1 Design of concrete structures – Part 1-1: General rules and rule for buildings
BSI
[16] COUCHMAN. G. H , HICKS, S. J and RACKHAM, J, W
Composite Slabs and Beams Using Steel Decking: Best Practice for Design & Construction
(2nd edition)
SCI P300, The Steel Construction Institute, 2008

41 / 42
[17] BS 8110-1 Structural use of concrete. Code of practice for design and construction, BSI,
London, 1997.
[18] BAILEY, C. G.
The influence of thermal expansion of beams on the structural behaviour of columns in steel
framed buildings during a fire
Engineering Structures Vol. 22, July 2000, pp 755 768
[19] EN 1993-1-8:2005 Eurocode 3: Design of steel structures – Design of joints
BSI
[20] LAWSON, R. M.
Enhancement of fire resistance of beams by beam to column connections
The Steel Construction Institute, 1990
[21] EN 1363-1:1999 Fire resistance tests. General requirements
CEN
[22] EN 1365 Fire resistance tests for loadbearing elements.
EN 1365-1:1999 Walls
EN 1365-2:2000 Floors and roofs
EN 1365-3:2000 Beams
EN 1365-4:1999 Columns
CEN
[23] RFS2-CT-2007-00042 : FICEB+ - Fire resistance of long span cellular beam made of rolled
profiles ; March 2011
[24] O. Vassart, Analytical model for cellular beams made of hot rolled sections in case of fire, PhD
Thesis, Université Blaise Pascal Clermont-Ferrand II, 2009
[25] NF EN 1991-1-2 : Eurocode 1 – Actions sur les structures – Partie 1-2 : Actions générales,
actions sur les structures exposées au feu.
[26] NF EN 1993-1-2 : Eurocode 3 – Calcul des structures en acier – Partie 1-2 : Règles générales,
Calcul du comportement au feu.
[27] MT ACB+ : ArcelorMittal Cellular Beams – Logiciel ACB+ - Descriptif Technique – CTICM
Rapport 7.072-01

42 / 42