EN 1991: Actions on structures – Part 1.

4: Wind Actions

only considered on those parts of a roof whose length exceeds the smaller of W/2 or H, and for walls
on those parts whose length exceeds the smaller of W or 2H. In EN1991-1-4, friction effects on walls
and roofs are only considered when the roof or wall length exceeds the smaller of 2W or 4H.
EN1991-1-4 also has a simple rule allowing friction effects to be disregarded when the area of the
surfaces parallel to the wind (i.e. roof and side walls) is equal to or less than 4 times the area of all
surfaces perpendicular to the wind (i.e. windward and leeward walls). The area over which the
friction force is assumed to act is smaller in EN1991-1-4 than BS6399-2, however EN1991-1-4 does
not allow the use of the size factor when calculating friction forces, these are counteracting effects
but there will be differences between friction forces calculated from these two methods.

2.4.6 Section 6 Structural factor cscd

This section gives rules for determining the structural factor cscd which accounts for the effect of non-
simultaneous occurrence of peak pressures over the surface of the structure or element combined with the
effect of dynamic response of the structure or element due to turbulence. For the majority of traditional low-
rise or framed buildings cscd may conservatively be taken as 1.0 . Clause 6.2(1) of EN1991-1-4 gives a list of
building and element types for which cscd may be taken as 1.0. For other building types, or where a more
precise value is required, cscd must be determined using the detailed procedure given in clause 6.3.1. Two
alternative procedures are given in Annexes B and C of EN1991-1-4 for determining the parameters required
for the calculation of cscd, the recommended procedure is given in Annex B. The UK NA allows Annex B but
not Annex C for use in the UK.

The factor cscd is generally treated as a single factor unless otherwise specified in the NA. In the UK it has
been decided to allow cscd to be separated into a size factor cs and a dynamic factor cd. This is similar to the
approach used in BS6399-2, where the size effect factor, Ca, corresponds to the size factor cs and the
dynamic augmentation factor, (1 + Cr) corresponds to the dynamic factor cd. The decision to allow cscd to be
separated was not taken just to promulgate the BS6399-2 approach but because of the positive benefits to
be gained from reduced wind loads and also because the procedure for determining cs and cd in Annex B is
complex and not suited for hand calculation. The NA gives Table NA3 for determining cs values and Figure
NA9 for determining cd. These have been derived using the detailed procedure in Annex B. In Figure NA9,
graphs are given for four classes of structure which correspond to the classes given in Figure 1 of BS6399-2.
The benefits of separating cs and cd are greatest for large plan area low-rise buildings. For example consider
wind loads on the long face of an office building 10m high with plan dimensions of 60m x 20m in Town
terrain, using the NA cd = 1.0 and cs = 0.77 giving a cscd value of 0.77. This will give a 23% reduction in wind
load on this face compared with the cscd value of 1.0 recommended in clause 6.3.1 of EN1991-1-4.

2.4.7 Section 7 Pressure and force coefficients

This section gives pressure and force coefficients for a wide range of buildings, structures and elements
(note that force coefficients for bridges and related structures are given in Section 8).

EN1991-1-4 uses four separate types of aerodynamic coefficient:

• Internal and external pressure coefficients

• Net pressure coefficients
• Friction coefficients
• Force coefficients

The first three will be familiar to users of BS6399-2. Force coefficients give the overall load effect on a
structure or element and include friction effects, whereas pressure coefficients do not. Therefore friction
coefficients may be applied in conjunction with pressure coefficients but should never be used with force
coefficients.

15
Guide to the use of EN 1991-1-4 – Wind Actions

2.4.7.1 Pressure Coefficients

Pressure coefficients are given for buildings in Section 7.2. There are many similarities between the
EN1991-1-4 and BS6399-2 pressure coefficients, because they were largely based on those given in the
BS6399-2 standard method. The layout of the tables and figures and the size of the roof and wall zones will
therefore be familiar to UK engineers. There are some omissions and differences, for example, EN1991-1-4
does not include directional pressure coefficients, information on inset stories, polygonal plan shapes, re-
entrant corners, irregular faces or multipitch roofs. The major differences are:

• Pressure coefficients
Two sets of pressure coefficients, cpe,1 and cpe,10 are given in EN1991-1-4 for most building forms.
cpe,1 values are intended to be used for the design of small elements and fixings, and cpe,10 values
2
are for design of loaded areas over 10m . In general the cpe,10 values are similar to those given in
2 2
BS6399-2. For loaded areas between 1m and 10m EN1991-1-4 recommends that logarithmic
interpolation be used using Figure 7.2. After consultation with UK industry it was concluded that this
would impose a considerable unnecessary burden in calculation effort. Consequently Figure 7.2 is
not to be used in the UK. The NA gives a simplified alternative procedure in which the cpe,1 values
2
should be used for all fixings and elements of 1m or less and the cpe,10 values should be used for all
2
loaded areas >1m .

• Division by parts
The rule on division by parts has been changed, it may now only be applied to the windward walls; in
BS6399-2 it could be applied to all walls but only for determining overall loads

• Pressure coefficients for vertical walls

In EN1991-1-4 these values depend on d/h ratio, this is the inverse of the ratio used in BS6399-2.
The effects of wind funnelling are not included in EN1991-1-4. Also note that Table 7.1 should not be
used in the UK. It has been replaced in the NA by Table NA4 which now includes net pressure
coefficients for overall load; this corresponds to Table 5a in BS6399-2.

• Pressure coefficients for circular (vaulted) roofs

Figure 7.11 pressure coefficients for vaulted roofs, should not be used in the UK. This has been
replaced in the NA by Figures NA10 and NA11 which provide more reliable data and include the
effects of roof length. Pressure coefficients for vaulted roofs are not given in BS6399-2.

• Internal pressures
In BS6399-2 only two possible values for internal pressure, +0.2 and -0.3 are given for the general
case with no dominant openings. It is known that the internal pressure can take values within this
range, or even outside of this range in some circumstances. The approach used in EN1991-1-4 is
based on the permeability of each wall and roof and allows a more precise value of internal pressure
to be calculated depending on the approaching wind direction.

2.4.7.2 Net pressure coefficients

Net pressure coefficients are given for canopy roofs, free-standing walls and parapets. These data are based
on BS6399-2, however, there are some differences, for example:

• Monopitch canopies
Some of the pressure coefficients have been increased and the distance of the centre of pressure
from the windward edge has been reduced from 0.3W used in BS6399-2 to 0.25d in EN1991-1-4.

• Free-standing walls and parapets

In BS6399-2 a reduction factor is given to account for end effects, in En1991-1-4 this factor has been
combined with the pressure coefficients. The figure for the effect of shelter from upwind walls has
been simplified.

16
 EN 1991: Actions on structures – Part 1.4: Wind Actions

2.4.7.3 Force coefficients

Force coefficients are given for signboards, structural elements, circular cylinders, spheres, lattice structures
and flags. BS6399-2 does not give any force coefficients; however’ it does give net pressure coefficients for
signboards and for a limited range of structural elements, which for these cases can be considered to be
equivalent to force coefficients. It should be noted that for slender structures with h/d > 5 the force
coefficients for elements may be used. This causes a discontinuity around h/d = 5 because the loads for
h/d<5 are derived from pressure coefficients and the loads for h/d>5 are derived from force coefficients. This
more likely reflects the uncertainty and lack of information on force/pressure coefficients on tall buildings
rather than an aerodynamic effect.

2.4.8 Annexes

There are six informative annexes in EN1991-1-4:

• Annex A - Terrain effects

• Annex B - Procedure 1 for determining the structural factor cscd
• Annex C - Procedure 2 for determining the structural factor cscd
• Annex D - cscd values for different types of structure
• Annex E - Vortex shedding and aeroelastic instabilities
• Annex F — Dynamic characteristics of structures

Of these Annexes C, D and E will not be used in the UK.

Annex A contains five parts of which part A.2 Transition between roughness categories 0, I, II, III and IV, will
not be used in the UK because transition effects are contained in Figures NA3 to NA8 in the National Annex.

Annex B gives the recommended procedure for determining cscd and is based on the procedure given in the
ENV version of EN1991-2-4.

Annex F gives recommendations for determining the basic structural parameters that are required for
determining dynamic response of structures, such as:
• natural frequencies
• modal shapes
• equivalent masses
• logarithmic decrements of damping.
Annex E does not include adequate information on aeroelastic response of bridge decks. For this reason
Annex E will not be used in the UK. It will be replaced by a new document based largely on the existing
Annex E but with additional information for bridges. This new document will not be part of EN1991-1-4, but is
expected to be published as a companion document (NCCI).

2.5 Road map

EN1991-1-4 includes two general procedures for calculating wind loads on buildings and structures, these
are:

a) a standard procedure that applies to those structures whose structural properties do not make them
susceptible to dynamic excitation. This procedure applies to the following classes of structure:

• Any building of height less than 15m

• Fa ade and roof elements which have natural frequencies > 5Hz

• Framed buildings with structural walls of <100m in height and whose height is also less than 4
times their in-wind depth.

• Circular chimneys with height <60m and whose height is also less than 6.5 times their
diameter

17
Guide to the use of EN 1991-1-4 – Wind Actions

b) a detailed procedure that applies to those structures which are likely to be susceptible to dynamic
excitation and which fall outside of the scope of the standard procedure.

It is anticipated that the standard procedure will be used for the design of the majority of buildings and
structures in the UK, including dwellings, low-rise buildings and medium-rise buildings. The detailed
procedure is likely to be used, or at least considered, in the design of tall buildings, chimneys, masts, long-
span bridges and other similarly wind sensitive structures.

The number of steps required in EN1991-1-4 to determine wind loads for the majority of buildings and
structures is the same as in BS6399-2. In general the 6 steps outlined in Table 4 are required (the equivalent
steps in BS6399-2 are also shown in this table for comparison).

Table 4 Summary of calculation steps to calculate wind load using EN1991-1-4 and BS6399-2
EN1991-1-4 Procedure BS6399-2 Procedure

1. Read Vb,map from map, adjust for altitude 1. Read Vb from map

2. vb = cdir cseason cprob vb,0 2. Vs = Vb Ss Sd Sa Sp

3. qb = 1/2 ρ vb2 3. Ve = Vs Sb

4. qp = ce,flat(z) qb 4. qs = 1/2 ρ Ve2

5. We = qp(z) cpe 5. p = qs Cpe Ca

6. Fw = cscd Σ W Aref 6. P = 0.85 (Σ Pfront - Σ Prear ) (1 + Cr)

Table 5 outlines the steps required to determine wind loads on buildings and structures where z < 50m and
orography is not significant. In other cases the number of steps increases, however the NA has greatly
simplified EN1991-1-4 by introducing alternative procedures for determining the design wind load by
presenting parameters such as the roughness factor cr, the turbulence factor (ki/ln((z-hdis)/zo), the exposure
factor ce,flat and the size and dynamic factors cs and cd in a tabular or graphical format. This speeds up the
design process, reduces the calculation effort and improves the accuracy over the EN1991-1-4 original
procedures, whilst still being fully compatible with the EN1991-1-4 principles. Figure 1 shows the flow chart
for determining wind loads using the NA and EN1991-1-4 simplified and general methods.

18
 The case study TSTs

Figure 1 Flow chart for determining wind loads using the UK NA with EN1991-1-4

Check Limits of
applicability 1.1(2)

Apply altitude factor to UK

Determine fundamental basic
windspeed map NA2.4
wind velocity vb,0 NA2.3
vb,o = vb,map calt

Calculate basic wind velocity, vb

Calculate basic
wind pressure qb

Find mean wind velocity,

vm(z) using Figures NA3 and
Find ceflat(z) using
NA4
Figures NA7 and NA8 (if
General Method
z < 50m and orography is
Find turbulence intensity, Iv (must be used if
significant find co(z) (A3))
Simplified using Figures NA5 and NA6 orography is
Method significant and
z > 50m)
Determine peak velocity Determine peak velocity
pressure using expression pressure using expression
NA4 NA3

Option 1: cscd = 1
Determine size effect factor cs
Option 2: Table NA3, Figure NA9
and dynamic factor cd
Option 3: 6.3.1 and Annex B

Find pressure coefficients (cp)

or force coefficients (cf)

Wind Loads

19
20
Table 5 Main differences between EN1991-1-4, including the National Annex, and BS6399-2

EN1991-1-4 BS6399-2 UK National Annex Commentary

Scope Buildings and civil eng. Buildings and their No change Much wider scope in EN, includes: buildings, bridges, chimneys, masts, towers, lattice structures, etc.
works components but excludes guyed masts, lattice towers, lighting columns

Methods Standard procedure Standard method Modified in NA 2.19 The standard procedure may be used for buildings where H<15m or 100m < H < 4D or for a wider
Directional method range of buildings using the modified procedure given in the NA

Detailed procedure Hybrid method The detailed procedure should be used for buildings which have a significant dynamic response

Wind speed Fundamental basic wind Basic wind speed Vb Map Figure NA1 EN1991-1-4 is based on a 10-minute mean wind speed, BS6399-2 is based on a mean hourly wind
Guide to the use of EN 1991-1-4 – Wind Actions

velocity vb . . speed
Mean wind velocity vm No equivalent Corresponds to vs So in BS6399-2 where it is only used in annex C.2.1
Basic wind velocity vb Site wind speed Vs The basic wind velocity corresponds to the site wind speed in BS6399-2
vb is calculated for the actual building height. Vs is calculated for a 10m reference height
No equivalent Effective wind speed Ve EN1991-1-4 does not directly calculate gust wind speed
Orography factor co Topography increment Sb Same as the procedure used in the BS6399-2 directional method
Altitude factor calt Altitude factor Sa Given in NA 2.4 Similar to BS6399-2 procedure but allows a reduction in calt with height above ground
Directional factor cdir Direction factor Sd Given in NA 2.5 Same as the BS6399-2 procedure
Season factor cseason Seasonal factor Ss Given in NA 2.6 Same as the BS6399-2 procedure
Roughness factor cr(z) Fetch factor ScTc Modified in NA 2.10 The NA gives a simplified procedure for determining cr(z) using Figures NA3 and NA4
Turbulence intensity Iv(z) Turbulence factor St Tt Modified in NA 2.15 The NA gives a simplified procedure for determining Iu(z) using Figures NA5 and NA6
2
Exposure factor ce(z) Terrain & building factor Sb Modified in NA 2.16 The NA gives a simplified procedure for determining ce(z) in flat terrain using Figures NA7 and NA8

Terrain Five terrain categories Sea, Country Town Modified in NA 2.10 The NA combines the five EN1991-1-4 terrain categories to give three categories that correspond to
categories 0, I, II, III and IV those in BS6399-2, Sea (cat 0), Country (cat I & II) and Town (cat III & IV)

Wind Peak velocity pressure Dynamic pressure Modified in NA2.16 The NA gives an alternative procedure for qp(z) which corresponds to the BS6399-2 procedure, except
Pressure qp(z) qs that the gust factor gt, which is a variable in BS6399-2, has a fixed value of 3.0

Pressure Orthogonal pressure Orthogonal and directional Modified in NA 2.24 The NA gives an alternative procedure for determining cpe for areas between 1m2 and 10m2. EN1991-1-
coefficients coefficients cpe,1 and cpe,10 pressure coefficients 4 does not include directional pressure coefficients.

Force Given for some buildings No force coefficients given EN1991-1-4 gives force coefficients for tall slender buildings, lattice frames and elements
coefficients and elements
 References

3. References

1. EN1991-1-4, Eurocode 1: Actions on structures – General actions – Part 1-4: Wind actions, CEN
2005.

2. BS6399-2: 1997, Loading for buildings – Part 2: Code of practice for wind loads, BSI, June 2002

3. Cook N J, Designers' Guide to EN1991-1-4 Eurocode 1 Part 4 - Wind actions. ISBN 0727731513
4. Rees J, The assessment of site wind speeds at Bilsdale Mast using computational fluid dynamics,
st
IASS Working Group 4 for masts and towers, 21 meeting, Milan, September 2003.

21