Professional Documents
Culture Documents
Li
Code of
I ~
This Code of Practice represents a standard of good practice and therefore Lakes the form of recomsnend2tions. Compliance with
it does not confer immtwirv from relei’ant statutor~ and Legal requirements.
2
CP3 : ChapterV: Part2 :1972
Foreword
Part I of the etirrent edition of this Code of Practice, which deals with dead att(I iiitposed IOa(l5 other than
wind loads, was published in 1967; Part 2 deals with wind loads.
In 1970 a revision of the 1952 edition was l)tlbliShCd as Part 2 in which pressure cooflicients and force
coefficients for a limited range of building shapes wcre given. The present edition extends this range by
the addition of lilessure coeflicients for all building shapes for ~vltiehthere is suflicient information available.
It also gives force coefficients for unclad structures and a(lditional appendices on ice formation and wind
forces on circular sections.
The committee responsible for the preparation of this Code appreciates that there are many building
shapes not covered, but the necessary data arc not available for these to be iflelu(led in a code and there is
need for additional research, particularly on the relationship between wind tunnel conditions and those in
reality, to enable figurcs to he given for furthet shapes.
As compared with the 1952 edition, this edition treats wirld loading in more detail because changes in
building (lcsign and construction have, for some types of buildings, increased the influence of wind loading in
relation to other imposed loads. This edition also uses wind data that have become available since 1952.
An important change in this edition from thc 1952 cdition is the adoption of gust loadings as the basis
for design in place of the mean load averaged over one min~ite, which had been the basis hitherto. There
is evidence to justify this change which will lead to a more realistic assessment of the XViII(l loads. At the
same time the opportunity has beeti taken to assess the probable wind spccds on a statistical basis in
accordance with the current advice of tlic Meteorological Office,
hi this Code numerical values have been given in terms of SI units, details of which arc to be found in
US 3763, The International System of units (SI) For the convcnicnce of those who wish to use them,
‘.
approximate equivalents are also given in metric technical units and in imperial units; mote accurate
conversions can bc ol)tained from uS 350, Conversion factors and tables A ~vindspeed and pressure
‘.
conversion chart (St metric imperial) is given in Pig. 3. For the purposes of this Code, the principal
(hifference between SI units and metric technical units lies in the measurement of force, load and load
intensity. The St unit of force and load is the newton (symbol, N); it is that force which, when acting on
a mass of I kg, imparts to it an acceleration of I rn/s2. Thus, under conditions of standard gravity, the
following relationships apply:
I kgf = 9.806 65 N (exact value)
I lbf 4.448 N (approximate value).
3
CP3 :ChapterV: Part2 :1972
1. Scope
This Part of this Code gives methods for calculaing the wind loads that sh6uld be takcn into account
when designing buildings, structures and components thereof.
~does not apply to buildings or structures that are or unusual shape or location for which special
i~stigations may be necessary to determine the wind loads and their effects. Experimental wind tunnel
data, where available, may be used in place of the coefficients giveil in this Code.
fluildings that may be subject to scriotis wind cxcitcd oscillations rcquirc further investigation. It should
be noted that such oscillations may occur at wind velocities lower than the maximum. -
NOTE. ‘Thc title of the Draft for Development referred to in this Code is given on the inside back cover.
2. DefinItions -:
For the purposes of this Code, the following definitions apply:
(I) )JreadtIa~. The dimension of the building normal to the direction of tlt~ wind.
(2) Dept/a. The dimension of the building measured in the direction of the wind.
(3) height. The height of a building above the ground adjacent to that building.
(4) Lengdst. The greater horizontal dimension of a building above the grot~nd adjacent to that building;
or the length, between..supports, of an individual,structural member,
(5) WIdThf. The lesser horizontal dimension of a building above the ground adjacent to that building;
he width of a structural member across thc direction of the wind.
(6) height above ground. The dimension above the general level of the ground to windward.
(7) Element ofsurface area. The area of surface over which the pressure coefficient is taken to be constant.
(8) Effectivefrontal area. The area normal to the direction of the wind or • shadow area’.
(9) DynamIc pressure of wind. The free stream dynamic pressure resultant from the design wind speed.
(10) Pressure coeflicient. The ratio of the pressure acting at a point on a surface to the dynamic pressure
of the incident wind.
(NI) Force coefficient. A non-dimensional coeflicient such that the total wind force on a body is the
product of tile force coefficient multiplied by the dynamic pressure of the incident wind tirncs an appropriate
area, as defined in the text.
(12) Topography. The naturc of the earth’s surface as influenced by the hill and valley configurations.
(13) Ground roughness. Thc nature of the earth’s surface as influenced by small-scale obstructions such
~str~ca and buildings (as distinct from topography).
JJreadzh end depth of a buihdiag arc dimensions related to the direction of mhc wind.
Length and width are dimensions related to the plan form.
4
CP 3 : Chapter \/: Part 2 :1972
3. Symbols
A clement of surface area
A, = effective frontal arca
b breadth
C1 force coefficient
normal force coefficient
transverse force coefficient
C1. frictional drag coefficient
—
C, es pressure coefficient
C,, external pressurc coefficient
internal pressure coefficient
d depth
D dianieter
F force
normal force
F1 transverse force
F’ — frictional force
h — height
H height above ground
] width of member as indicatcd in diagram
Jo. width of member across direction of wind
k m a constant
K reduedon factor
I length
p pressure on surface
p. — external pressure
= internal pressure
5
CP 3 : Chapter V: Part 2 :1972
4.2 It is important to note that the wind load on a partially completed structure will be dependent on the
method and sequence of construction and may be critical. It is reasonable to assume that the maximum
design wind speed 1”, will not occur during a short construction period, and a reduced factor 83 can be
used to calculate the probable maximum wind. The graphs of Fig. 2 should not, however, be extrapolated to
a period of less than two years.
Pressure coefficients and force coefficients are given in 7 for a range of building shapes. Forec coefficients
are given in B for unclad structures.
6
CP 3 : Chapter V: Part 2 :1972
that is likely to be exceeded on the average only onec in 50 years. Table I gives the basic wind specds~ to be
used in some cities and towns in the United Kingdom. For other locations the basic wind speed should be
obtained by interpolation between isopleths shown in tho map in Fig. I evaluated to the nearest integral
metre per second.
5.2.2 It should be assumed that the basic wind may blow from airy horiaontal direction.
Table 1. Basic wind speed in metres per second for some cities and towns
ni/s rn/s
Aberdeen 49 Leeds 46
Bedford 40 Leicester 44
Belfast 45 Liverpool 46
flirkenhead 46 London 38
Birmingham 43 Londonderry 52
Blackpool 46 Luton 40
Bournemouth 41 Manchestcr 45
Bradford 47 Middlcsbrough 45
Brighton 40 Newcastle-upoa-Tync . 45
Dristol 43 Northnrnpto~ 42
Cam bridge 41 Norwich 43
Cardi 1[ 45 Nottingham 45
Carlisle 46 Oxford 40
Coventry 44 Perth 51
Dumfries 47 Plymouth 44
Dundee Portsmouth 40
Edinburgh Preston 45
Exeter 43 Sheffield 46
Glasgow 51 Southampton 40
Great Yarmouth 43 Stoke 43
I luddcrsfield 46 Swansea 46
Inverness 49 Wolyerhampton 43
I pswich 41 York 45
Kingston-upon-I-lull ‘45
These values apply to cities ~nd towias only and not necessarily to the surrounding areas.
Cha nitel Islands:
Alderney 52
G tierusey 52
Jersey 52
• An explanatory note (Appendix B) shows how ¶hcsc spveds have bec~ estimated and gives the basis for ietccth~g the
3-second gust speed.
7
CP 3: ChapterV: Part 2 :1972
NA Nil NC
At’ NE OA
52
OF
NF NJ
4~
OL
NA. NM ~ ~,. AI~RDEEN
If
7~UNI’EE
PEUFI ~/
BAN ~ ~—
I
NO N NU 00
0
OINBURGI 46
GLASGOW
ov ow
it NY EWCAS rLE
LON DOND
~‘ ‘jAY
BELrASY
• ~ ~ TB
SE
N-....
(.
P~2~STOr4 L E ED5\o
C ULL
N
1G
2 - LIVE P0
- SH~
STOKE NOTTIH(IIAFA
0
4
44
so’ NOItWIO4o
ej~~.” LEIC ESTER
siW~~ I OIITMINGII NM
1AMPTOt
50 AaE YSTWYTII ~FDFORO# 42
NOR TI
.~
~. .•~.••
SEA
~RDlFF/
I su OXFORD
3IPSWIcH
2 / I—-
SQ DR lSTO 1. 38
I--,
DO L)RIAIi NA OU
I IRIGIATON
SW ~
PLYMOUTH ~42
~S•.•~4 4 TV
(
~-46
50 48 —413
1,0 50 6Onds
p t A I I I I I I I I
8
CP 3 : Chapter V: Part 2 :1972
5.4 Topography factor S~. Examination of the wind speed records has shown that the height’ of the site
above sea level does not itself affect the value of the basic speed, so unless special local effects are present
the value of S1 should be taken as 1.0.
Exposed hills rising well above the general level of the su~rounclin~ terrain may give rise toaccelerated
winds as may some valleys, particularly those that are so shaped that funnelling of the wind occurs with
wind blowing up the valley. Sites so affected are often well knoxvn locally for their abnormal winds. For the
above-mentioned sites values of S1 of 1.1 should be used..
• On the other hand there are some steep sided, enclosed valleys where wind speeds may be less than
normal. Caution is necessary in applying a reducing factor but for such eases a value of ‘~& may be used
for S1.
• Values of the topography factor S1 are given in Table 2 to cover most conditions. Local knowledge may
help the designer to select the S1 value, but if he is in doubt on the matter the advice of the Meteorological
• Office should be sought. Addresses of the relevant advisory offices arc given in Appendix H.
Topogrnphy VnIIIC or s,
5.5 Ground roughness, building sire ~‘~idheight above grou;~d, facter ~2• The factor S~ takes account of the
combined effect of ground rou.~liuiess, the variation of wind speed with height above ground and the size
of the building or component part undcr consideration.
In c6nditions of strong wind, the \vifld speed usually increases with height above ground. The rate of
increase depends on ground roughness and also on whether shart gusts or mean wind speeds are being
considered. This is related to building 5i7.C to take account of the fact tha small buildings and elements
of a building are more affected by short gusts than are larger buildings, fe~ which a longer wind-averaging.
period is more appropriate.
5.5.1 Ground roughness. For the purposes of this Code the ground ro ~ucssis divided into four categories
and buildings and their elements are divided into three clan;~ as fot’ ws
Ground roughness I. Long fetches of open, le~’el or nearly 1ev’ country with no shelter. Examples are.
flat coastal fringes, feris, airfields and grassland, moorlaud or F railand without hedges or walls ardund
tli6 fields.
Grot~nd roughness 2. Flat or undulating country with obstruct ions such as hedges or walls around fields,
scattered windbreaks of trees and occasional buildings. Examples are most farmland and country estates
with the exception of those parts that arc well wooded.
9
CP 3 : Chapter V: Part 2 :1972
Ground roughness 3. Surfaces covered by numerous large obstructions. Examples are well-wooded park-
land and forest areas, towns and their suburbs, ~ind the outskirts of large cities. The general level of roof-
tops and obstructions is assumed at about 10 m, but the category will include built-up areas generally apart
from those that qualify for category 4.
Ground roughness 4. St~rfaccs covered by nuijierous lar~c obstructions with a general roof height of about
25 m, or~ more. This category covers only the centres of large towns and cities where the buildings are not
only high, but are also not too widely spaced.
Further explanation of these ground roughness categories will be found in Appendix A.
5.5.2 Cladding and building size. The wind speed fluctuates from moment to moment and can be averag~d
over any chosen period 6f time. It has been found that the shortest time scale, 3 s, that is normally measured
produces gusts whose dimensions envelop obstacles up t.o 20 in acro~s. The longer the averaging time the
greater is the linear length encompassed by tbe gust. For this reason three classes have been selected as
described below.
C’hiss A. All units of cladding, glaziiig and roofing and their immediate fixings andl individu~d members
of unclad structures (see 7.2, Note). •
Class B. All buildings and structures where neither the greatest horizontal dimension nor the greatest
vertical dimension exceeds 50 in (165 ft).
Class C. All buildings and structures whose greatest horizontal dimension or greatest vertical dimension
exceeds 50 m (165 ft).
The values of S2 for variation of wind speed with height above gr6und for the various ground roughness
categories and the building size classes are given in Table 3.
The height should be taken to the top of the structure or, alternatively, the height of the structure may
be divided into convenient parts and the wind load on each part calculatedl, using a factor S2 that corresponds
to the height above ground of the top of that part. The load shouldi be applied at the mid height of the
structure or part, respectively. This also applies to pitch roofs.
Further explanation of the relationship between g~sts and wind speed averaging time is given in
Appendix B.
In the previous paragraphs height above ground means the dimension above the general level of the
ground in the vicinity of the building, assuming that there are no unusual conditions. Allosvanees should,
however, be made for special conditions. For example, the height above ground of a building on a cliff
top or a steeply-sloped escarpment should be taken as the dimension above the foot of the cliff or escarpment
for winds from that direction. Details of the rule to be follo~ved are given in Appendix D.
5.6 Factor S3. The factor 53 is basedl on statistical concepts, y2,iven in Appendix C, which take account of
the diegree of security required and the period of time in years dlunug which thcre will he exposure to wind.
Whatever wind speed is adoptedi for diesien plirpelses, there is always a probability, however small, that
it may be exceeded in a storm of exceptiorlal violence; the greater Jie period of years over which there
~vill be exposure to the Windi, the greater is t Iris prc.haluihity. Fig. 2 shows values of S~ equivalent to a period
of exposure of 50 years plotted against this period of years.
Normally wind loads on completed structures andl buildings should be calculated at S~ I with the
following exceptions.
(I) Temporary ~.truetures.
(2) Structures where a shorter period of exposure to the wind may he expected.
(3) Structures where a longer periotl of exposure to the wind may be required.
(4) Structures where greater tlurn normal safety’ is reqiriredi.
For these special cases both the exposure to the wind an(l the probability level may be varied
neriocl ol~
accord~ng to circUmstances. In n~ ci :~c js the period of cx pesn re to ta l:cu as less than 2 years. Some
examples arc givcn in Appeitdix C’.
10
CP 3 Chapter V: Part 2 1972
Table ~. Ground roughn~ss, building size and height above ground, factor S2
(t) Open Country with; (2) Opeii couuiutry wi(h 0) Country with ninny (4) Sairrace winiwge
no ohstrtictioiis scattered windt)rcaks wmtthrcatcs smalt towns; nod frequent ohstruictions,
outskirts or large cities c.~. city cchtres
3 or less 0.83 0.78 0.73 0.72. 0.67 0.63 0.64 0.60 0.55 0.56 0.~2 0.47
5 0.88 0.83 0.78 0.79 0.74 0.70 0.70 0.65 0.60 0.60 0.~5 0.50
10 1.00 0.95 0.90 0.93 0.88 0.83 0.78 0.74 0.69 0.67 0.62 0.58
15 1.03 0.99 0.94 1.00 0.95 0.91 0.88 0.83 0.78 0.74 0.69 0.64
20 1.06 1.01 0.96 1.03 0.98 0.94 0.95 0.90 0.85 0.79 0.75 0.70
30 1.09 1.05 1.00 1.07 1.03 0.98 1.01 0.97 0.92 0.90 0.85 0.79
40 . 1.12 1.08. 1.03 1.10 1.06 1.01 1.05 1.01 0.96 0.97 0.93 0.89
50 1.14 1.10 1.06 1.12 1.08 1.04 1.08 1.04 1.00 1.02 0.98 0.94
60 1.15 1.12 1.08 1.14. 1.10 1.06 1.10 -1.06 1.02 1.05 1.02 0.98
80 1.18 1.15 1.11 1.17 1.13 1.09 1.13 1.10 1.06 1.10 1.07 1.03
1.20 1.17 1.13 1.19 1.16 1.12 1.16 1.12 1.09 1.13 1.10 1.07
100
120 1.22 1.19 1.15 1.21 1.18 1.14 1.18 115 1.11 1.15 1.13 1.10
140 1.24 1.20 1.17 1.22 1.19 1.16 1.20 1.17 1.13 1.17 1.15 1.12
160 1.25 1.22 1.19 1.24 1.21 1.18 1.21 1.18 1.15 1.19 1.17 1.14
180 1.26 1.23 1.20 1.25 . 1.22 1.19 1.23 1.20 1.17 1.20 1.19 1.16
200 1.27 1.24 1.21 1.26 1.24 1.21 1.24 1.21 1.18 1.22 1.21 1.18
‘11
CP3 :ChapterV:part2 :1972
200
I..
‘V
41)
I-.
tn
0
Ii
0
-o
.2
Factor S3.
Fig. 2. A statistical factor S~
12
CP 3 : Chapte1 V: Part 2 :1972
q k~’
=
Values of/c arc as follows for thc various units used in this Codc:
k 0.613 in SI units (N/rn2 and mis)
k 0.062 5 in metric technical units (kgf/n9 and mjs)
k 0.002 56 in imperial units (lbf/ft2 and mile/h).
Any vertical line traversing the scaics on Fig. 3 indicates approximately the direct convcrsion between
velocity and pressure on all three systems of units.
10 6.3 7.6 9.0 10.6 12.3 14.1 16.0 18.1 20.3 22.6
20 25.0 27.6 30.3 33.1 36.0 39.1 42.3 45.6 49.0 52.6
30 56.3 60.1 64.0 68.1 72.3 76.6 81.0 85.6 90.3 95.1
40 100 105 110 116 121 121 132 138 144 150
50 156 163 169 176 182 189 196 203 210 218
643 225 233 240 248 256 264 272 281 289 298
70 306
13
C~ 3: ChapterV: Part 2 :1972
30 2.3 2.5 2.6 2.8 3.0 3.1 3.3 3.5 3.7 3.9
40 4.! 4.3 4.5 4.7 5.0 5.2 5.4 . 5.7 5.9 6.!
5.0 6.4 6.7 6.9 7.2 7.5 7.7 8.0 8.3 8.6 8.9
60 9.2 9.5 9.8 10.2 10.5 10.8 11.2 11,5 11.8 12.2
70 12.5 12.9 13.3 13.6 14.0. 14.4 14.8 15.2 15.6 16.0
80 16.4 16.8 17.2 17.6 18.1 18.5 18.9 19.4 £9.8 20.)
90 20.7 21.2 21.7 22.1 22.6 23.1 23.6 24.1 24.6 25.1
100 25.6 26.1 26.6 . 27.2 27.7 28.2. 28.8 ‘29.3 29.9 30.4
110 31.0 31.5 32.1 32.7 33.3 33.9 34.4 35.0 35.6 36.3
120 36.9 37.5 38.1 38.7 39.4 40.0 40.6 .41.3 41.9 42.6
13 43.3 43.9 44.6 45.3 46.0 46.7 47.3 48.0 48.8 49.5
1 50.2 50.9 51.6 52.3 53.1 53.8 54.6 55.3 56.1 56.8
57.6
14
CP 3 : Chapter V: Part 2 :1972
Cu
U)
Ci)
Cu
0.
C,
E
Cu
•~0
-o
C
Cu
0)
0)
U)
I-
0
•4~.
Cu
1)
C
0
U)
L.
Cu’
C
0
U
‘-mr-----1
i(ti~OlOA
~tw eu =rj 0)
15
CP 3 : Chapter V: Part 2 :1972
)When calculating the wind load on individual structural elements such as roofs and walls, and individual
‘~adding units and their fixings, it is essential to take account of the pressure difference between opposite
faces of such elements or units. For clad structures it is therefore necessary to know the internal pressitre
as well as the external pressure. The following distinguishing pressure coefficients are therefore used:
(I) external pressure coefficient, C~~;
(2) internal pressure coefficient, C’1~.
The load F acting in a direction normal to the individttal structural element or cladding unit therefore is
F (C,,,—C~1)qA *
—
where A is the surface area of the structural element or cladding unit. A negative value Ibr 1”’ indW~slcs that
the resultant force is outwards.
Values of C1, are given in Tables 7 to IS and a method of assessing the ~‘alucsof C1,1 is given in
Appendix E.
7.3 Force coefficients. Force coefheients, when given, apply to a building or structure as a whole, and
~vliemimultiplied by the effective frontal area A~ of the building or structure and by the dynamic pressure q
give the total wind load on that l)articular building or structure thus:
F C,qA,—
where F is the force acting in a direction sl)ecifled itt the table and Cr is tlte force coeflicient for the bttilding.
h n~c design wind speed and tltcrcfore q (tepentis on S~, which varies with height (see 5.5.2).
16
CP 3 : Chapter V: Part 2 :1972
It should be noted that the value of the force coefficient differs for the wind acting on different faccs of
a building or structure. In order to determine the critical load, the total wind load should be calculatcd
for each wind direction.
If the wind load is calculated by dividing the area into parts, the value of C’1 applied to each part should
be that for the building as a whole.
7.4 Frictional drag. In certain buildings of spccial shat~ a force due to frictional (Irag should be taken in
addition to those loads calculated from 7.2 and 7.3. For rectangular clad buildings this addition is necessary
only where the ratio 11 or 11.ts greater than 4. The frictional drag F’ in the direction of the wind is given
— —
ii b
by the following.
If h ~ b, F’ Cpqb(d—410.1- C,.q2h(d—4h), or
if Ii ~ Li, F’ Cr.qb(d4b)+C,.q211(dilb).
The first term in each case gives the drag on the roof and the second the drag on the walls. The terms are
givell separately to allow for the use of different valnes of C,, and q on the ditrerent suFes.
C,. = 0.01 for smooth surfaces without corrugations or ribs across the wind direction.
C,. 0.02 for surfaces with corrugations across the wind direction.
C,. 0.04 for surfaces with ribs across the wind direction.
For other buildings the frictional drag will be indicated, where necessary, in the tables of pressure coe0i-
cicnts and force coefficients.
17
CP 3 : Chapter V: Part 2 :1972
Table 7. Pressure coofficients C~0 for the walls of rectangular clad buildings
NOTE. I, is Ibe height to eaves or parapet, 1 is the greater horizontal dimension of a building and w is the lesser horizontal
dimension of a building.
‘3
OP 3 : Chapter \/ : Part 2 :1972
~ LIF ~
60 +0.2
±0.5 —0.5
—0.5 —0.8
—0.8 —0.7
—0.7
NOTE I. A is ~hchcight to cavcs or parapet and w is the lesser horizontal (IjmCIlsiofl of Key plan
a building. y y
NOTE 2. The pressurc cocflicicnt on the undcrsidc of any roof ovcrhnng should bc
~‘cn as that on ~hcadjoining wall surfacc.
Hh— 1h
)fhcrc no local cocfflcients arc given ~I~eo~’ciaII cocflicicnls apply. F E C
Wind
a: F H /
w
— It or 0.15w,
19
CP 3 : Chapter V: Part 2 :1972
—I
L~y
Wind ,11 ~
A
‘I lI~; ___
I,
—I
NOTE. Arca 11 and area L refer tothe whole quadrant.
0~-.
“~ -U
2
-a’- -o
5 —1.0 —0.5 —1.0 —0.9 —1.0 —0.5~~ —0.9 — 1.0—0.5 —1.0 —2.0 —1.5 —2.0 —1.5 —2.0 —2.0
10 1.0 —0.5 —1.0 —0.8 —1.0 —0.5 —0.8 —1.0 —0.4. —1.0 —2.0 —1.5 —2.0 — 1.5 —2.0 —2.0
15 —0.9 —0.5 —1.0 —0.7 —1.0 —0.5 —0.6 —1.0 —0.3 —1.0 —1.8 —0.9 —1.8 1.4 —2.0
—
—2.0
20 —0.8 —0.5 —1.0 —0.6 —0.9 —0.5 —0.5 —1.0 —02 —1.0 —1.8 —0.8 —1.8 1.4 —2.0 —2.0 —
25 —0.7 —0.5 —1.0 —0.6 —0.8 —0.5 —0.3 —.0.9 —0.1 —0.9 —1.8 —0.7 —0.9 —0.9 —2.0 —2.0
30 —0.5 —0.5 —1.0 —0.6 —0.8 —0.5 —0.1 —0.6 0 —0.6 —1.8 —0.5 —0.5 —0.5 —2.0 —2.0
NOTE. I, is the height to eaves .~i lower side, I is the greater horizontal dimension of a building and ii’ is the lesscr horizontal dimension
of a building.
20
CP 3 : Chapter V: Part 2 :1972
raNo 10. Force coefficients C~ for rectangular clad buildings with flat roofs
(acting in the direction of the wind)
~4dk.
Wind
— FIT. ~4 1.2 1.3 1.4 1.5 1.6
LJi~i
<~ 0.7 0.7 0.75 0.75 0.75
—F
..*IclH A
•2
LII1~
‘
H—d—.-j
T
b 11 0.9 0.95)1.0 1.05 1.1)1.2 1.4
NOTE. b is the dimension of the building nornial to the wind, d is the dimension of the buildink ~neasuredin the dircctioa
of the wind, I is the greater horizontal dimension of a builiIin~ and w is the lesser horizontal dimension of a building.
.21
CP 3 : Ch’apte,r V: Part 2 :1972
~t,.
¼)
~
yzh or 0.1w
Roof plan whichever is
the lesser
~
Sect ion
degrees degrees
5 0 —0.9 —0.6 —0.4 — Q.3 —0.3 —0.3 —0.3 —0.3
‘30
45
—0.2
±0.3 —0.6
—0.6 —0.4
—0.6
—0.3
—0.4
—0.2
—0.2
—0.3
—Th4
—0.2
—0.2
—0.5
—0.5 I
Distar~cc
Roof angle Wind angle (I Il[ /12 /13
degrees degrees
Up to 45 90 —0.8 —0.6 0.2
Frictional drag: when wind angle ~ 0~ horizontal forces due to frictional drag are allowed for ~n the
above values;
when wind angle ~ — ‘JO0 ‘illow for frictional drag in aeeord:tnee with 7.4.
NOTE. Evidcrlce on these buildiags is fraginent~mry and any departures from tile cases given should bC investigated
separately.
22
CP 3 : Chapter V: Part 2 :1972
/%.
WI I WI I W
—I—
Reof
Section
degrees ——.-————---— — -
Distance
Wind nngte a 1I~ Jig /13
degrees
90 —0.8 —0.6 —0.2
Frictional drag: when wind angle ~ — 0~ horizontal forces due to frictional drag arc allowed for in the
al)ove values;
when win(l angle ~ 900 allow for frictional drag in accordance with 7.4.
NOTE . Evidence on these buildings is fragmentary and any (JeJ)artIIrCs from the cases given should be investigated separately.
23 -
CP 3’: Chapter V Part 2 :1972
1~
Table 13. Pressure coefficiepts C fc~ r canopy roofs with -~ ~ —i < 1
Section 1Zoof~aiigle
Windward slope Leeward slope
deg t-ces
5- • 0.S or +1.2 —1.0
10 —0.6 or -1-1.4 — 1.0
15 ~0.4 or +1.6 -1.0
20 -
.0.2 or +1.8 —0.7
25 0 or±2.0 rn02
30 0 or +2.0 0
‘The coefficients take account of the combined effect of the wind Ofl both upper and lower surfaces of the
canopy. Where alternative coefficients are given the canopy should be designed to accept both loading
conditions. •~n addition to thc uplift .forces,therc ~vill be horizontnl loads on the canopy due to th~ wind
pressur~ on ihy fa~ci~and to wind friction over, the surface ~fthe roof. For any wind dircctioh, only one
~ these two, the more onerous, need he ~akeii into ~iccouiit, Fascia loads shouhl, he calculated on the area
Ar the surf~tce facing the wind, Lisiuga for
coefficient of.1.3.
- )Frictional drag should be calculated using the coefficients given in 7.’1.
C’~, for individual cladding panels should be takcii as ±2.0.
For ~nonopitch canopies the centre of pressure should be taken as acting at ‘0.25 of the span from the
windward cdge. For double pitch canopies the centre of pressure should be taken as acting at themid—point
bf each slope.
24
CP 3: ChapterV: Part 2 :1972
Table 14. Force coefficients C for clad buildings of uniform section
(acting in the direction of the wind)
Cr (or hcIgtt/bre~dth ratio
See also Appendix G Smooth ~ 6 0.5 0.5 0.5 0.5 0.5 0.6 0.6
[~~~1g Euipse • < 10 0.5 0.5 0.5 0.5 0.6 0.6 0.7
.~94. b/d 1/2 ~ 10 0.2 0.2 0.2 0.2 0.2 0.2 0.2
- b/d~. I
f—i.——
ED
<10 0.7 0.8 0.8 0.9 1.0 1.0 1.3
b/d 1
rib 1/6 10 0.5 0.5 0.5 0.5 0.6 0.6 0.6
0~
<3 0.3 0.3 0.3 0.3 0.3 0.3 0.4
b/d= 1/2
T
rib — 1/2
0.2 0.2 0.2 0.2 0.3 0.3 0.3
~iiL ~5
bid
rib =
1/2
1/6
All
values
0.5 0.5 0.5 0.5 0.6 0.6 0.7
b/d~2 All
~ rib 1/12 values 0.9 0.9 1.0 1.1 1.2 1.5 1.9
25
CP3:ChapterV:Part2:1972
ThbIe 14 (cominuod)
I)
26
I
CP 3 : Chapte1 V: Part 2 :1972
0
ThbJe 14 (continucd)
Where, strakes are used, b may be taken as tlic breadth over the strakes. Structures that, because of their
size and the design wind velocity, arc in the supereritical flow regime may need furthcr calculation to
casure that the grcatcst loads do not occur at some wind speed below the maximum when thc flow will
bc subcritical.
The coefficients arc for buildings ~vithou~ projections, except where otherwise shown.
In this tablc J’
8b is used as an indication of the airflow regimnc..
1 In coursc of prcparamion.
27
~jCP3:ChapterV: Part2 :1972
WI.~d
—9-
Pressure coeffIcient C
~ 0 Surfiicc: rough or wIsh projections Surriace: smooth
For the purpose of calculating the wind forces that act in such a way as to deform a cylindrical structure
the values of C~4 in tablc 15 may be used. They apply only in supercritical flow (i.e. they should only be
used where D > 0.3 in). They may be used for wind blowing normal to the axis of cylinders having thcir
axis normal to the ground plane (i.e. chiinncys, silos) and to cylinders having their axis parallel with the I)
ground plane (i.e. h6rizontal tanks) provided that the clearance between the tank and the ground is not less
‘han D.
<)his the height of a vertical cylinder or length of a horizontal cylinder. Where there is a free flo~v of air
around both ends, 1* is to be taken as half the lengtl~ when calculating hiD. Interpolation may be used for
intermediate values of h/D.
lit the calculation of the load on the periphery of the cylinder, the value of C~1 shall be taken into
account.
For opcn ended cylinders where h/D ~ 0.3 G~ titay be taken as —0.8.
For openended cylinders where hi/D < 0.3 ~‘~1may be taken as -—0.5.
28
CP 3 ; Chapter V: Part 2 :1972
8.1.1 Icing. Icing will affect the shape an4 loading on the structure, but at present it is thought unlikely
to occur with extreme winds. Such information as is at prcscnt known is given in Appendix F.
8.2 Force codflicients o[ Individual members. The coefficients refer to membcrs of infinite length. For
members of finite length the coefticiciits should be multiplied by a factor K that depends on the ratio IlL.
where I is the length of the member and j~ is the width across the direction of the wind. Values of K are
given in Table 16.
Where any member abuts onto a plate or wall in such a way that free (low of air around that end of the
member is prevented, the ratio IlL should be doubled for the purpose of determining K. When both ends
of a member are so obstructed, the ratio should be taken as infinity.
Table 16. Values of reduction factor K for members of finite length and
sicn d e rn ess
I/hurl/i) 2 5 10 20 ‘10 50 100
Circular cylinder,
/ subcritical flow 0.58 0.62 0.68 0.74 0.82 0.87 0.98 1.0
N
Circular cylinder.
supc~critical flow 0.80 0.80 0.82- 0.90 0.98 0.99 1.0 1.0
Flat plate perpendicular
to ~~imid. . 0.62 0.66 0.69 0.81 0.87 0.90 0.95 1.0
29
CP3:ChapterV:part2:1972
8.2.1 Fla-sided nmenmbers. Thc force coeflicients in Table 17 are givcn for two mutually-perpendicular
ditcetious relative to a referemmec axis omi tlme structural rueniber. Timey are designated ~ and 6’~~ and give
L the forces normal and transverse, respectively, to the refercimec plane as will be apparent from the diagrams.
Force coefficients are for wind normal to the longitudinal axis of the member.
• Normal force F~ C,~qKlj.
Transverse force F1 C~1qK II.
Table 17. Force coefficients C,~ and (‘fe for individual structural members
(flat sides) of infinite length
NOTn. In timis table the force cocllicknm G~ is given jim rctation to time dinmcnsinn J amid not, as in odmcr cases, in rchmtio:in to lime
ctfcctivc frontal arcs ,f,.
I)
3Q
CP 3 : Chapter V: ~~art
2 :1972
8.2.2 Circular sections. For circular sections, the force coefficients which arc dependent upon values
~,
of DV,, are given in Table 18. The valttes of Ce given in this table are suitable for all surfaces of evenly
distributed roughness of height less than 1/100 diameter, that is, for all normal surface finishes, and for
nicmbers of infinite length.
Force, F C~qKID.
6 ~ DV, 12 m2/s
<
45 =DV, < 90 ft
mile/li 0.6
4.1 x l0~ ~=Re < 8.2x i0~
- A description of supcrcriical how and Reynolds number (Re in abovc tablc) is given in Appendix G.
8.2.3 Wires and cabks. For wires and cables the force coefficients c~ that arc dependent upon values of
• DV, arc, given in Table 19.
Table 19. Force coefficients Ct for wires and cables (I/V > 100)
Flow re’glemic Force coctilcient C, For:
moderately smoolb
smooth surface wire wire (galvanized or fine stranded cables thick utrat3dcd cables
painted)
— .—•. —-—————-..•—.. ——————-— ———..‘-e
DV, < 0.6 m2/s (4.5 ft mile/li) .— 1.2 1.3
31
CP3:ChapterV:Part2:1972
4
1.3 SlagIc rram~. Since the wind can conic from any direction the most unfavourable load condition
shouLd be taken. In general, the wind load on a single frame should be calculated for the condition where
the wind is at right angles to the frame unless it can be shown that another wind angle is appropriate.
The wind load acting on a single frame should be taken as
ctrcular accilons
5oJidlty ratio ~5 hat-skied member,
uut)crItICnI flow supercrltlcal flow
DV. <6 m2/s (45 ft mile/h) D V. ~‘ 6 m2/g (45 ft mile/h)
32
CF 3: C~)apier V: Part 2:1972
• ‘~ ~.4 ~ rranie strueturc~.’litis section applies to structttrcs havitig two or more parallel frames whcre
hc windward frame tnay have a shielding eflect upon the frames to lccward. The windward frame atid
any uttshcltcred parts of other frames should be calculated as in 8.3, but the wind load on the parts of
__ fratucs hat aic shcltered should he multiplied by a shielding factor :j, which is dependent upon the solidity
ratio ol the windward frame, the type of memt)er comprising the frame and the spacing ratio of the fratnes.
The values of t lie shielding factor arc givcn in Table 21.
__ \Vhtere there arc more than two frames of similar geometry and spacing, thc wind load on the third and
subsequent fratnes should be taken its cqual to that on the second frame.
The loads on the various frames should be added together to obtain the total load on the structure.
iL Table 21. Shielding factor’j
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.3 and ovcr
~ ———— .—
it[) to 1.0 1.0 0.96 0.90 0.80 0.68 0.54 0.44 0.37
2.0 1.0 0.97 0.91 0.82 0.71 0.58 0.49 0.43
$L) 3.0
4.0
1.0
1.0
0.97
• 0.98
0.92
0.93
0.84
0.86 -
0.74
0.77
0.63
0.67
0.54
0.59
0.48
0.54
5.0 1.0 0.98 0.94 0.88 0.80 0.71 0.64 0.60
6.0 and over 1.0 0.99 0.95 0.90 0.83 0.75 0.69 0.66
The spacing ratio is equal to the distance, cetitre to centre, of the frames, 1)eatns or girders divided b~
the least overall dimension of the frame, beam or girder measured at right anglcs to the direction of lb
wind. For triangular framed structures or rectangular framed structures diagonal to the wind the spacin
ratio should be calculated from the mean distance between the frames in the direction of the wind.
The aerodynamic solidity ratio used in Table- 21 enables all cross sections of single members to b
tncorporatcd.
Acrodynwuic solidity ratio /7 = solidity ratio (~) x a constant
8.5.1 Lattice towers ol’ square and eqitilateral (riangttlar section constitute special cases for which it txi~
I be convenient to use an overall force coefficient in the calettlation of wind load. The wind load shoul
br convenience, be calculated for the condition when the wind blows against any face.
The ~viitdload F acting in the direction of the wind should be taken as
F C~ q~l~
where ,l, is the ciThetive area of the face (see 8.3);
1~~ q is the dynamic pressure of the wind (see 6); and
C’, is the overall force coeflicicat (sec Tables 22, 23 and 24).
( —
33
i
h CP 3 : Chapter V: Part 2 :1972
±1 8.5.2 For towers with flat-sided members, the values of the overall force coefficient are given in Table 22.
ii 0.1
0.2
3.8
3.3
3.1
2.7
0.3 2.8 2.3
0.4 2.3 1.9
0.5 2.1. 1.5
For square lattice towers the maximum load occurs when the wind blows onto a corner. It may be taken
as 1.2 titnes the load for the face-on wind.
• For triangular lattice towers the wind load may be assumed to be constant for any inclination of the
wind to face.
8.5.3 Since it is only in very few cases with lattice towers composed of members of circular cross section
that all the members of a lattice tower are in the same flow regime, i.e. either subcritical or supereritical,
wind force calculations should be carried out as descri bed in 8.3 for single frames, due account being taken
of the shielding factors given in 8.4.
When it can be shown that all members of the tower are wholly in the same flow regime, the overall
force coerncients q given in Tables 23 and 24 may be used. These tables are based on.actual measurements
and Live somewhat lower values than would be obtained using Tables 20 and 21.
Table 23. Overall force coefficient Ct for square towers composed of
rounded members
Force cocflicient C, for:
Appendix A
Ground roughness, building size and height above ground: the basis of the
S, factors
A.1 Ground roughness—the four categories. Near the ground the wind encounters various obstacles in its
pathand the gustiness of the wind depends on the size, frequency and geometrical arrangement of thcse
obstacles. If the obstructions are large and frequent the surface over which tho wind flows is said to be
rough; if the ground surface itself is level and there are no obstructions then the surface is said to be smooth.
A rough surface will produce a wind flow with much gustiness whereas a smooth surface will not add
appreciably to the gustiness already developed in the lower layers of the atmosphere. The gustiness can be
defined numerically in a number of ways but the one adopted for the purposes of this Code is tlte ratio
of the maximum gust, 1’, to the mean hourly wind speed F, both having a probability of 0.02 of being
e~cecdcd only once in any one year. This ratio has been determined for all ancmograph sites in the United
Kingdom. Thc ratio V/V is found to vary from about 1.25 to about 2.4. U~ing the known site characteristics
it is possible to classify site roughness into four categories each of which has an associated mean ratio V/V
A.2 Variation with height. Above an effcctive height of 10 m it is sufficient to assume that both mean
hourly speeds and maximum speeds averaged over any selected intcrval of time between I h and 3 s vary
with height according to a power law
‘~i= (~X
where V~, is the maximum speed averaged over a giv~n interval of time at height H znetrcs,
V is the maximum speed averaged over the same interval of time at height 10 in and
c is the exponent for the power law that is specified for each averaging time and roughncss category.
11
When 10 m, (~)~becQJnes unity for all values ot’~, and V~ V.
shown
however,
that the
special
general
experiments
level of the
using
rooftops
three may be takeit at
anemometers intothe
consideration
Post Office and the inpower
Tower law have
London modified.
also
Suppose general rooftop level is Y metres and VE is the maximum speed averaged ovet an £tpprOl)riate
dine interval at a height of 10 in above 1’, the power law becomes
(If
Il—I’
—~ ___
10 /
A.3 Wind speed averaging times. It is shown in Appendix 13 that maximum wind speeds of a stated return
period and averaged over periods of time equal to 3 s, S s and 15 s are required for design purposes.
llccause the anemograph in widespread use cannot be analysed to extract S s and iS s maximum speeds,
the results of sl)ccial experiments have to be used to obtain the relationship between the maximum gust
and the maximum speeds averaged over 5 s and 15 s intervals of time. It has been found that on the average
• at ;tn effective height of F metres the relationsltip is:
S s speed (f’5) = 0.95 x 3 s speed (f”3)
9 ISs speed (V15) = 0.90 x 3s speed (f”3).
35
is
~j CP3 : ChapterV: Part 2 :1972
Introducing these factors the maximum speed averaged over I seconds can be obtained at height ii metres
front a knowledge of the maximum speed averaged over 3 s at a height of Ii metres because
)
A.4 EffectIve height. Finally, whatever the terrain, it has been assumed that the niaxintutn speed averaged
over any time interval is the same at gradient wind level. Allowances for the variation of gradient wind
height with roughness of terrain cannot be made exactly but using the power law formula it is possible
to calculate speed ratio at different heights and the factor (.~) has been introduced such that at certain
heights, different for each ground roughness category, the 3-second gust speed is constant whatever the
roughness of the underlying terrain.
It should be noted that Y represcnts a level corresponding to general rooftop or obstruction level. Thus
in Category 1 it ~ assumed that the general obstruction height is zero, itt Category 2 the nican level of
obstruction is about ~ m, in Category 3 the mean level of obstruction is about 10 m and in Category 4 the
obstructions have a general mean height of 25 m. Category 4 is thus seen to represent the centres of densely
packed cities only where the buildings are tall (25 in or so) and the streets relatively narrow.
Thc cifective height B is 10 m above the general rooftop or obstruction height Y.
1.
For the 3-second gust these rattos are:
(I) Speed at 12 at in Category 2 0.98 .speed at 10 in in Category I.
(2) Speed at 20 m in Category 3 0.95 speed at 10 in in Category I.
(3) Speed at 35 in in Category 4 0.94 speed at 10 in in Categorj I.
Introducing these ratios
ft where V10, ~ V, the basic 3-second gtis~ at 10 m height in open country (map speed).
Defining S2 as the number by which the maximum speed averaged over a time of 3 s at a height of 10 m
above open level country (i.e. the basic speed) should be multiplied in order to obtain the maximum speed
averaged over, a, time of I seconds at 1! metres above the different terrain where the general level of
4’
obstructions (or rooftop level) is Y metres,
02~7jI~y,J~ 10 P
Collecting the above factors and adding the values for the exponent, the values adopted front experiments,
measurements and trial and error methods are:
( E. (V~
2_\ (Vrn\
Category (metres) (metres)
kv~, S ~‘.VIEU) a
: (seconds) I (seconds)
3 5 15 3 5 15
36
I—
CP 3 : Chapt~.1 V: Part 2 :1972
is the conversion factor to ensure that at or near gradient wind height all 3 s speeds are thc
• is the ratio of the : second to the J-sccond wind at an effective height E, and
7’ ~ is the selected power law exponent appropriatc to the gust averaging time and the roughness
LL. categories.
Th~ airflow near the ground between the obstructions will be a composite of wakes, deflections and
channehlings whose character cannot at present be codilicd owing to lack of obscrvational data and a
systematic theory. The pressure corresponding to a 3-second gust at 3 m height was fixed arbitrarily. The
9 •. prcssurc at 10 m above general obstruction level was defined by the procedure described above. It has
been assumed that the pressure increased linearly from the fixed value at 3 in to the defined value at 10 in
‘I. above getieral obstruction level. From these definitions S2 factors have been calculated.
A.5 Influence of change of roughness. Although classification of ground roughness into Categories I to 4
has been made, it should be recognized that the change from one ground roughness to another is necessarily
-~ a gradual process. The wind must traverSe a certain ground distance before equilibrium is established in a
new velocity profile. The change starts first in the layers of wind nearest the ground and the new profile
extends to an increasingly deep layer as the fetch increases.
For practical purposes it may be assumed that a fetch of a kilometre or more is necessary to establish
a different roughness category, but that within the actual roughness layer, i.e. below the general rooftop
or obstiuction level to windward, a lesser distance mayapply as follows, depending on the density of buildings
and other obstructions on the ground:
Crourut covorace Rcqulrcd fctch
it • -12 10% 500 ni
-1230% lOOm
• -1215% • 250in
Shichter due to an individual building or obstruction should itot be considered without expert guidance.
For a site where the ground roughness is different from different directions, the most severe grading
L . shotihl be used, or, exceptionally, appropriate gradings may be tised for different wind directions. For
example; the seafront of a coastal town would generally rank as ground roughness Category 1 (see also
V
-l
Appendix 13
Explanatory notes on gusts and the wind-speed averaging time
The natural period of oscillation of most structures is only’ a few seconds or less amid, since impulsive forces
2 • lasting only about a half-period of oscillation are effective iii deflecting a structure, it follows that gusts
of only a few seconds duration would produce significant wind loads if they developed simultaneously over
the whole structure.
The incidence and spread of gusts over a building has been studied at the Iluilding Research Station
where it has been found that the gust loading on a tall rectangular ofliec block is significantly greater than the
nic~in load averaged over one minutc as used in the 1952 edition of the Code.
Measurements suggest that the time interval over which maximum wind speeds should be averaged will
dcpend on the size of the building or part of a structure under consideration. This averaging time varies
from about IS s for buildings of 50 ni (165 ft) height or horizontal sl)read to I s or less for small elements
such as windows, cladding usiits and roof coverings.
37
I
CP 3 : Chapter V: Part 2 :1972
01
As a routine, at the netwom K of .inemograph stations maIntained by the Meterological Ollice in the United
L1 Kingdom, the mean hourly wind speed and the maxinmumii gtmst of the day arc extracted from the record.
cause of the characteristics of the anemometer, the niaximnumn gust speed represents the mean wind speed
• averaged over nl)out 3 s. It is said to have an averaging time of 3 s and is often called the 3-second gust.
~ • Similarly the mean hourly wind speed has an averaging timile of one hour.
• Re1~itively little is known about the incidence and intensity of gusts averaged over I)eriOds of time shorter
than 3 s, 50 that, although it is desirable to determaimic maximum wind loads on small units using gLists
with shorter averaging times, it is not at present possible to specify mnagnititdes in Strong winds. The standards
I adopted in the Code are, therefore, the nmximnuni 3-second gust speeds for all units of glazing, cladding
and roofing, whatever the size or proportion of the building concerned. A winch speed with 5 s averaging
J is
time used for ‘the structural design of buildings and structures where neither the greatest horizontal nor
vertical climiiension exceeds 50 ni (165 ft). For buildings and structures whose greatest horizontal or vertical
dimension exceeds 50 m (165 ft) a wind speed of IS s averaging time is used.
-or 115 stations in the United Kingdom, it is possible to analyse the series of annual maximum gust
~ speeds statistically following the methods of Gumbel. A similar j)rocedure is also adopted for tIme atmalysis
of a series of annual maximum mean hourly wind speeds. However, in order to prepare a map of extreme
I values, the data must be homogeneous and refer to a standard datum level in a standard site. Extreme
values for other sites and at other levels may then be estimated by the use of factors.
The basic wind speeds given in Table I and on the map in ni/s have been adjustedas necessary from the
j ~esults of the statistical computations. The maximum 3-second gust speed likely to be exceeded on the
• average only once in 50 years was chosen because
(I) of the ease for selecting the 3-second gust given above, and -
(2) the average lifetime of most buildings covered in the Code is near 50 years.
L It should be noted that a vahie likely to be exceeded on the avcrage only 6tmce in 50 years also has a
probability of 0.02 of being exceeded omily once in any year and is often called the 50 year value or the
value having a return period of 50 years. An attempt has been mnadd to enable the designer to choose speeds
• of different return periods by the use of suitable factors.
1.
• Appendix C
A statistical factor S~
Factor S3 is based cmi statistical concepts and this appendix shows how the factor has beemi computed from
the data amid .lescribes sonic of the uses.
The wimid speech havimig a return period, 7; of 50 years should niore properly be called that speed which
will be exceeded just once with a probability p 0.02 in any one year. Return period an(l probability are
‘ • •
Now the probability that a value less than or equal to ~vindspeed x will occur in one year is q I —p
In a period of N years the probability Q that a value less tItan or equal to x will occur is
Q~q N
The probability 1’ that a value greater than x will occur at least once in a period of N years is
P=ml—Q
1— (1 p)N
For N = 50 and T =50, P 0.63. Therefore, there is only a probability of 0.63 that the once in 50 year
wind speed will be exceeded at least once in a period of 50 years.
38
j CP3:ChapterV:Part2:1972
V Three cases are considered as shown in Figs. 4, 5 and 6.
6
S
7777’rn7m
A?
C
C
A
Fig. 4. Slope of escarpment (1)
The artificial base Z. is that of the ground immediately around the building.
‘-C
(3) tan 0 ~ 2.
40
CP 3 Chapter V: Part 2 ‘1972
Appendix E
Conversely, openings at positions that arc experiencing external suction will reduce the pressure inside
the buildings thus increasing total loads on a windward wall, as shown in Fig. 8.
41
(I
The distributed permeability should be assessed in cacti case as accurately as is practicable. As a guide
can be said that the typical permeability of a house or oflice block with all windows nominally closed is
it
the range of 0.01 % to 0.05 ~ of the face area, depending on the d~gree of draughtproofing.
in
\X’lmcre it is not possible, or is not considered justified, to estimate the value of C~1 for a particular case,
the coefficient should be based on one of the following paragraphs for any determination of ~vattor roof
loading.
(I) Where there is only a negligible l)robabilitY of a dominant opening occurring during a severe storm,
C~ should be taken as the more onerous of -1-0.2 and —0.3
(2) For situations where a (lomninant opening is likely’ to occur, C1,1 should be taken as 75 % of the value
of Ci,, outside the opening. The extreme conditions should be determined for the various wind directions
that give rise to critical loadings and it ‘should be noted that especially severe internal pressures may be
developed if a dominant opening is located in a region ot’ high local external pressure.
There is a further coin p1 icat ion ii a wall or roof clenien t that comulirises severn I layers. For example, a
roof may be hoarded rind fel ted aid covered with tiles. The pressure difference bet~veen outside a4id inside
wilt then be broken (lowIm uito st & ps, across cad I layer; t liese steps will (lcpdii(t 6n the relative pcriiieabili y
oft lie various layers and the access of air to the spaces betwech diem. Each case needs careful study to ensure
that the whole of the wind load is not accidentally transfcrred to a single memnbramie such as a tf~a metal
sheet that may not be designed to carry it.
42
CP3:ChapterV: I~rt2:1972
Co,iirol of internal pressure. The value of C~1 can sometimes be limited or controlled to advantage l)y thc
• deliberate distribution of permeability in the walls and roof or by the deliberate provision of a venting
(levice that can serve as a dominant openii~g at a position having a suitable external pressure coefficient.
An example of such an application is a ridge ventilator on a low pitch roof which, under all directions of
wind, will reduce the uplift force on the roof.
Appendix F
Ice formation on structuros
F.! In winter or early spring each year at some place in the British Isles ice is deposited on structures and
the designer would ideally like to know
(I) what weight of ice can form on structures,
(2) what is the shape of (he ice dcposit,
(3) what is the density of the ice deposit,
• (4) what wind speeds arc likely during and after ice deposition, and
(5) how frcquently do these condhioums occur.
This appendix provides general guidance on icing conditions within the limitations set out below. In
the discussion mimean speeds are given because it is the build.up of ice that is important. An estimate of the
maximum gust speed when the ice load has been established can be made by multiplying the appropriate
mean speed by 1.5. The conditions leading to and after ice formation arc not likely to bc the same as those
in ~vhichiextreme gusts occur, so, ifextrcnie gusts calculated by the method described in 5 are used to compute
wind loads on iced structures an overestimate will be produced. Also, in strong winds icc will be blowum off
the structure and this may imiduce vibrations. No attempt has been made to discuss thesc vibration effects.
Occasionally two or perhaps all three types may occur simultaneously or in sequence and this possibility
is discussed in FAO. The wind forces or movements of the structure can be sufficiently strong to cause lumps
oh an otherwise even coating of ice to break off. Although, throughout the past 30 years or so, there have
becti reports of ice deposits in the meteorological literature, most of the knowzi occasions have been
descriptively but not quantitatively analysed and as a result reliable statistics cannot be cc~mpiled. This
appendix will, therefore, necessarily be fragme ntary.
V.3 ‘lIme most dense ice deposits arise xvhmen raindrops cooled to below 0~C fall on an exposed structure
and, perhaps after some running on the structure, freeze to forni a clear ice (leposit culled glazed frost.
‘[lie precipitation is known itS freezing rain or freezing drizzle. Reports of freezing precipitation are often
localized nnd occur once every few yeats iii some parts of time country, being mainly confined to England
amid Wales. The mnost widespread glazed frost of recent years, pcrhap.s of this century, occurred in January,
is
1940, and described by Drooks and Douglas*. They reported deposits of 1488 g of ice on a spray of beech
twigs weighing 100 g and deposits of 50 mnm diameter on telegraph wires. Another widespread glazed frost
occurred between II and IS March, 1947, but on this occasion the thickness of the deposit was generally less
well reported. Vertical surfaces exposed to freezing precipitation are generally coated with ice on the side
facing the wind. If chic surface isIlat and broad (e.g. a house side) the deposit has the form of a sheet of
ice of immure or less urjiform thickness, rarely ‘10 miii Co 50 miun thick. lf the surface is curved and relatively
BROOKS. C. 11. P., and DOUGLAS, C. K. M. The ~t ted frost of January. 1940. II.M.S.O., Gcophysical Akmoir No. 9t(.
43
CP 3 : Chapter V: Part 2 :1972
(
flAffOW laterally (e.g. trees, telegraph poles) the deposit tends to build smoothly, but with a thicker deposit
directly tnto wind rind thinner deposits at the sides of time object producing a change of curvature. Thicknesses
of up to 50 mar have been observed. If tIme surface is markedly curved and very narrow (e.g. cables, ~virc
rncsh fenecs) then the deposit, before freezing, immay run to completely encase the exposed surface often
• crnbcdding tIme support in tIme centre of thc ice section but just as frequently building an oval section ice
deposit with the wire roughly at the focus on the major axis furthest from the oncoming wind. If a cable
or wire is at an angle to the vertical or if it is horizontal, the asymmetry of tIme load may induce a twisting
moment. The coating oh’ ice may then exhibit spiral effects with a very uneven surface but the absence of
• spirals cannot be taken to imply that twisting has not taken place. Glazed frost adhmeres strongly to amost
• mctahlic, mineral or organic surfaces and is relatively dense; in the absenec of mneasurcmcnts it must be
assumed that time dcnsity is 0.92 g/cm3, thc density of purc water ice at 0 0C, because little or no air is trapped
in the deposit (when air is trapped the deposit is white).
• PA There are two typcs of meteorological situation in which the conditions for glazed frost may be niet,
• One is a steady situation in East or South-East winds with a narrow band Qf warm air overlying a very
cold surface layer of air aimd with very cold air above so that snow falhiimg from above is melted in the warm
layer and then time drops or droplets arc supercooled as they pass through the cooler underlying layer.
~ This situation may persist for days and is mainly reported in England and Wales. Time Imourly mean wind
‘~ speeds during formation are usually in the range from 6 rn/s to 10 nm/s arid, exceptionally, speeds of IS mn/s
may be expericimeed. Time glazed frosts of January, 1940, Marcim, 1947, and March/1969, are typical examples
in which deposits of 50 mm or so wcrc recorded on trees, cables and house sides, More frequently, time
meteorological conditions are met for only a short timc,after a cold spell. On these occasions a period of
or 3 ii of freezing rain may be followed by very strong warm winds with mean hourly speeds of 20 mu/s
• to 25 mIs, but melting of the ice takes l)lace rapidly. Such aim occasion was rcportcd on 4 March, 1970.
!Iecausc time deposit is formed in a short period of rain, thicknesses of 25 mm or so are unlikely to be
• cxceeded except on rare occasions. This type of glazed frost, but within variations in deposit timickinmess, is
rcportcd in some part of time Urmited Kingdom every year with’ wind speeds of up to about 15 nm/s or so,
mainly in Scotland but particularly on hills. While no detailed observations are available it is reasonable
to assume that temperature and precipitation conditions at heights up to 200 m or so above ground will
not vary greatly froni conditions at the surface. However, it may also be asstinred that wind speeds will
incrcase with height according to a power law with exponent 0.17 for hourly mean speeds. Thus an hourly
• mean wind speed of 15 rn/s at the surface beconies 2$ ni/s or so at 200 m.
In Scotland or on hill tops in England and Wales another, equally likely criterion might be:
Maximum ice timickness • 15 mm to windward
• Ice density 0.92 g/cmn3
Hourly mean wind speed during forniation 15 rn/s at the surface irmereasing according to time power
law with exponent 0.17
Hourly mean spced afler 2 or 3 Ii 20 mn/s at the surface increasimig according to tIme power
law with exponcrmt 0.17.
These two sets of criteria spring from different meteorological situations and the probability of their
occurrence cannot be assessed at time puesent time.
44
CP 3 : Chapter V: Part 2 :1972
F.6 The least dense ice deposits are formed when fog or cloud is blown, usually at steady hourly mean
speeds of 3—S m/s, onto cold surfaces, freezing on ininpaci to forni a loose aggregate feather ice or rime ice
that builds into wind in the shape. of pointed icicles. Air is trapped between the frozeim droplets that are
~vhite in appearance. The dcimsity of rime ice has beeu measured in coimtrohled conditions by Macklin
~vlmostudied the influence of wind spced, ambient air teniperature, droplet diamneter and liquid watcr
concentration, but no densities havc been determined so far as is known fromn actual atmospheric samples.
• Mackim found densities of from 0.1 g/cm3 to 0.9 g/cm3 beiimg deperindent directly omi droplet diameter
and wind speed but inversely proportional to temminperature of thmc aceretirig surface. For practical purposes
a value of 0.5 g/c;min’~ may be adopted because tIme very dense eases are associated with large droplet diameters.
The ice is usually fragile, disturbaimees in the airflow and movements of the depositing surface being of
sufficient force to result in lumps of ice breaking off. Jim steady wind cominditions, on fairly solid objects like
muasts, rime ice will forum a banimer kilo i/me wind. If time object is thin enouglm, e.g. wires, there will be only
one bainmer but flat narrow objects may acquire two banners, one at cachi edge facimig the wind. These
• banners immay grow to lengilins of 600 nimt. Occasionally time banners may ‘fill the spaces between adjacent
• upright ninembers of a structure, time strength of time ice deposit being increased by the additioimnl support
and tIme temmdency to fracture being much reduced.
1.7 Duriimg the period November to March in the British Isles rime ice may be exj)erienced on structures
rising frommin ground at 200 ni or so above sea level, particularly on tIme windward slopes of hills. Usually
banners of less than 200 mimi length are recorded at lower levels but banners of 600 ninmmin or more may occur
at ltX)0 iii or so above sea level every year and ormee every few years at lower altitudes. However, tIme lengtlm
t)fthmC banner will depend on tIme size of tIme depositing surface, its shmape~ the steadiness, particularly in
dircctioim, of tIme wind and time duratioim of the conditions of iciimg.
V.8 It is suggested tlmat for this type of deposit, suitable design criteria might be:
Maximum icc length into wind 600 mum
• Deimsity of ice 0.5 g/cm3
The hourly mean speed during and after fornination - 10 nm/s at 200 imi above sea level risiuing to
30 in/s at 1000 m above sea level.
These criteria should be applied for all structures on hills at lincighits greater than 200 ni above sea level.
l”,9 Ice mminay also be deposited as melting snow omi cold ~urfaces.Time resulting deposit is whitish in appearance
• and is nmidway betweenin glazed frost aimd rime ice botlm in appearance and density, time latter being about
• 0.7 g/cnm~’ if time one mmmcasu’remmmeimt of 38 lb/ft3 ninade on ice from tIme cables of the television mast at Emley
Moor, Yorkshire, in March, 1969, is accepted. Because it is a midway case no design criteria are offered
but hourly immeaim wind speeds of 1$ in/s are fairly common, particularly imm Scotlamind.
V.10 Commmbinatioims of tIme three iminethods of dcpQsition in any order are possible but perhaps the sequence
at Eumley Moor during March, 1969, will serve to illustrate time complexities. IJuring die period 12—19 March,
1970, hourly mucan speeds varied from 3 mini/s to 9 Ill/s. ‘throughiotit thm~ period fog was reported at telminpera-
lures near Doe to about —2 0C. Superiamposed on this mnmore or less continuous accretion of rime ice were
• periods of melting snow amid freezing rain. On the I 9th of March ice up to 160 mum in thickness was measured
omm cables of about JO immimm dianicter. On thm~s occaswmi, however, hourly menu wind speeds did not eAcced
10 mn/s even on time I 9dm when rapid thawing took place Within aii’ temperatures betweemi 1 amid 3 0C, timeO~
45
U
CP3:ChapterV:Part2:1972
For design purposes for lattice masts it is suggested that time criteria should be:
(I) Wall of ice filling tIme spaces bct’veen upright
members •
Apporiciix G
Wind forces on circular sections
Tlmc wind force on tiiy- object is given by
F~ C,.A,.q
• wimere C, is the forcccoeffmcient (sec 7 and 8)
• A, is the effective arca of the object normal to tIme wind direction, and
• q is the dynamic pressure of the wind (see 6).
For most shmapcs the force coefficient remains approximately constant over the whole rarmge ol’ wimid
speeds likely to be encountered. However, for objects of circular cross section it varies cominsiderably.
For a circular section tIme force coefficient depemids upon the way in which tIme wimid flows aroumind it amid
• is dependent upon tIme velocit5’ aiind kinematic viscosity of tIme wind amid diameter of the scetiomi. TIme force
coeffinciemint is usu~mIly quoted agaiiinst a non-dimensional parameter, c~ilhcd tIme Reymmolds imummiber, wlmichn
takes accoimmit of the velocity atind viscosity of time flowiming miniediuinini (iii this case the wiimd) and the inneniber
diameter.
DV
• Reynolds number, Re
1’
• whcrc D is time diameter of the member,
V, is the design wind speed, and
v is tIme kinematic viscosity of the air, which is 1.46 x iO-’~ mum/s at 15 0C and stamidard atmmmosplmcric
pressure.
Since hi most natural environments likely to be found iii time United Kingdom the kinematic viscosity
of the air is fairly constant, it is convenient to use DV, as the parameter iimstead of Reynolds numimbers amid
thuis has bcen donc in this Code of Practice.
Time dependence of’ it circular section’s force coefficient upon Reynolds mmumimber is dtine to time change in
time wake developed behind the body.
At a low Reynolds n4mmlber time wake is as shmo~vn in Fig. 9 and the force cocflicieimt is typically 1.2. As
Reynolds umumber is immereased tIme wake gradually chainges to timat sinowiin in Fig. 10, i.e. tIme wake width ‘1~
decreases and time separation point, s, moves from the front to tIme back of time body.
V —p.-
46
,
K.
CP 3: Chapter v: Part 2 :1972
PI~ As a result tIme force coefficient slmows a sudden drop at a critical value of Reynolds number, followed
by a gradual rise as Reynolds imuninber is increased still I’urtliner.
Tl~e variation of G~ with tIme h)aramcter DV, is slinown in Fig. II for infinitely hoimg circular cylinders hmaviimg
various values of relative surface roughness (cID)’, when subjected to a wind having an intensity and scale
of turbulence typical of built-up urban areas. The ctinrve for a snmootlm cylinder (<D I x I0~) in a steady
airstream, as found in a low-turbulence wind tunnel, is shown for comparison.
It can be seen thmat the main etfect or free-stream turbulence is to decrease the critical value of the
paranmeter DV,. For subcritical flows, turbulence can produce a considerable reductioim in ~ below tIme
steady airstream values. For supereritical flows, thinis effect becoimmes significantly snialler.
1]
71
ii
it
CE
‘a
• For values of ~see Dai’a sliecis of ihic En~ineerium~ Sciences IXmia Unji, 25h—259 Rcgeni Sired, London WI R 7AD.
I 47
1
I
CP 3 : Ch~ij~t~r V: Part 2 :1972
U)
-C £ a)
(I)
4) —
0)
4’)
~ L/)
a
Co
U
a)
Cl’)
0
0)
CTin
a)
a)
U)
0
Co
0)
Co
-c
CI)
a)
C.)
Co
U
i-.,
E c
\—
=A> 0
0
A.-
CC
a)
a)
‘4-
C)
C-)
am
0
0
4-
0
Co
0
U.) C) in 0)
I ci
48
I __
P3 ;ChapterVI’~irt2 ‘1972
Pagc
,rcword ‘ 3 • 32. Pressure coeflicicz4s C’~. for saw-tooth Page
Scopc ‘~ ~‘ . ‘ 4 roofs of iumulti-span buildings (mmli spans
Ddlnitions t;.:’~. . • 4 equal) with 1~4- w’
Symbols 5 23
13. Pressure coeflleiculs C~ for canopy roofs
~‘ 1’~~’ • ‘ .
Dynamic pressure of time wind 13 . -. l4~i3orce coeflicient~ C’, for clad buildings of
Prcs.suro coefficients and force coefficients 16 .‘ -“~uniformmm section (mmcting in the direction of
Force co~fflcients for unclad structures 29 wind)
•
25
15. Pressure distribution around cylindrical
structures
ppondicos 28
• 16. Values of reduction factor K for members
Ground roughness, building ‘size and •
oJ• finite lenglim and slenderness 29 6
above ground: time basis of tIme S3 17. Force coefficients C,~ and C,1 for individual
35 • structural mcnubers (flat sides) of i~finitc
Ezplanatory notes on gusts ~itd the wind- •
lcngtlm
. .
.
30
speed averaging time 37 • . . 18. Force coefficients Cr for individual struc-
A statistical factor 5’3 . . 38 •. . tural nuembers of circular section and
Tue effect of a cliff or escarpment on the . . . .. infinite length •, 31
height above ground Ii.. 39 • . .19. Force coefficients C’, for wires and cables
Thc estimation of internal pressure cocffl- (I/D>-100) 31
cients K • 41 . 20. Effective force coeflicients C, for siim~lc .
Values of
of’q in
q ‘in metni~
SI units teclmnical LillitS
(Nfm .‘ Figures ‘ ‘
Force coefficients C~ for rectangular clad Internal prcssure coefficient (2) •~.... >‘ :. 8. . .. 41
buildings with flat roofs (acting in time Wake in subcritical flow . . 9. 46
direction of the wind) 21 Wake in supereritical flow
• . ‘ • 10. 46
Psesature coefficients C!,,, for pitch roofs of Variation of force coefficient C, willi DV, . II.
inulil—span buildings (all spans equal) . for inflimitely long circular cylinders having
with Ii 4-Practice
Codi~ of WI represents a stimimclurd of good practice
• 22 iand therefore
various
kes the rorimi ofofrccorn,nenihitlovis.
degrees suiface roughness Cosiiplhinc.e 48
~dIt~
2.
U