Wind Loads on Long Low-rise Buildings

J. D. Ginger1 and J. D. Holmes2

1) Cyclone Structural Testing Station, James Cook University Townsville, QLD 4811, Australia
2) Department of Mechanical Engineering, Monash University, Clayton VIC, 3800, Australia

1. INTRODUCTION steep roofs. The revised AS/NZS 1170.2 [1]

Long low-rise buildings with spans greater than attempts to overcome this shortcoming by
30m and lengths exceeding 100m are often used increasing the magnitude of pressure coefficients
for bulk storage of materials such as sugar, on the leeward slopes of steep roof buildings.
cement and minerals. The structural systems of
such large buildings generally consist of portal or This paper summarises results from a wind tunnel
pin-jointed frames, sometimes spaced evenly at study carried out on a range of large buildings
the mid section and closer together at the gable- with gable and curved roofs. More details of
ends. Metal sheet cladding is attached to roof theses results are given by Ginger [7].
purlins and wall girts, which are fixed to these
frames. Cross bracing between the end frames 2. WIND TUNNEL TESTS
resist longitudinal (ie. in direction of ridge-line) The wind tunnel tests were carried out in the 2.0m
wind loads. Steep roof pitches of 30o to 50o are high  2.5m wide  22m long Boundary Layer
most common, but other low pitched and curved Wind Tunnel at the School of Engineering at
roof shapes are also used. In Australia, these James Cook University. Tests were carried out on
buildings are often located adjacent to port the four building configurations shown in Figures
facilities in tropical-cyclone prone coastal 1 to 4 and described in Table 1, at length scales of
regions, where wind loading is the dominant 1/200 and 1/300.
structural design consideration. Design wind
loads on the cladding and structural system of Building configuration Nos. 1, 2 and 3 were
these large buildings are usually determined using tested in a simulated rural boundary layer (Terrain
data given in wind load standards such as Category 2 in [1]), while configuration No. 4 was
AS/NZS 1170.2 [1]. tested in simulated urban terrain (Terrain
Category 3 in AS/NZS 1170.2).
An early study by Howe [2] in smooth uniform
flow identified the significant effect of aspect External pressures on the wall and roof panels
ratio (length, b / span, d) on wind pressures at the were obtained for approach wind directions () of
ends of low rise buildings. More recently, the -90o to 90o at intervals of 15o. Pressure taps were
variation of pressure distribution with varying connected to Honeywell pressure transducers via
aspect ratio on such buildings has been studied in Scanivalves and a calibrated tube and restrictor
boundary-layer flows, by Kanda and Maruta [3], system. The pressure signals were low-pass
Holmes [4] and Ginger et al [5]. Kanda and filtered at a frequency of 250Hz, and sampled at
Maruta [3] found that an increase in aspect ratio 500Hz for 24secs for a single run. The pressures
resulted in a variation in the area averaged were analysed to give mean, standard deviation,
pressure on windward and leeward slopes of the maximum and minimum pressure coefficients as;
roof. This effect was most noticeable for high C = p ( 12 U h ) , C =  p ( 12 U h ) ,
2 2

pitch roofs, on the leeward slope, for oblique p p

approach winds. Holmes [4] and Ginger et al [5] 
C ˆ = pˆ ( 2 U h ) and C  = p ( 12 U h )
1 2 2
showed that design wind loads on these buildings p p
was significantly underestimated by the previous
where, 1
2
U h 2 is the mean dynamic pressure at
edition of the Australian Standard AS 1170.2 [6],
especially for large aspect ratio buildings with height h, the mid roof level for the planar roof
buildings and maximum roof height for the 5. CONCLUSIONS
curved roof building. Wind tunnel studies were carried out to determine
pressure distributions and wind load effects on
The results were obtained from averaging the data long low-rise buildings with a range of roof
from five separate runs. The correlation shapes and varying aspect ratios. The following
coefficients between pressures on each pair of conclusions were reached:
panels on selected frame tributaries on some
building configurations were also determined. • Large mean and peak suction pressure
coefficients measured on the leeward half of
the roof and wall over a distance of b/3 from
3. PRESSURE DISTRIBUTIONS the windward gable-end, of the steep pitch
The peak (positive on windward half and negative gable roof buildings for oblique approach
on leeward half) pressure coefficients for  = 45o, winds (ie.  = 45o) increase in magnitude with
on the tributary of the second frame from the increasing aspect ratio (Figures 5, 6).
gable-end (ie. Frame B) on building • There is less variation with aspect ratio for
configurations 1, 2, 3 and 4 of varying aspect buildings with a low pitched gable roofs and
ratios are shown in Figures 5, 6, 7 and 8 curved roofs (Figures 7, 8).
respectively. Comparison of these peak values • The 1989 edition of the Australian Standard
with the corresponding peak pressure coefficients AS 1170.2 underestimated wind loads on
Cpeak derived from the previous AS1170.2 [6] for steep pitch gable roof buildings of aspect ratio
 = 0o, (Ginger [7]), shows that peak suction greater than 3 on areas near the windward
pressures were being significantly underestimated gable-end, compared with wind loads derived
on the leeward roof and wall on the steep pitch, from wind tunnel tests. This anomaly has
planar roof buildings of aspect ratio greater than largely been rectified in the 2001 edition.
3.
REFERENCES
4. STRUCTURAL LOAD EFFECTS 1). Standards Australia/Standards New Zealand.
A previous study by Holmes [4] showed that Structural design – General requirements and
structural load effects in supporting frames of a design actions, Wind actions AS/NZS 1170.2-
long building with a 36 degree gable roof, were 2001.
being grossly underestimated by the 1989 2). J.W. Howe. “Wind pressure on elementary
Australian Standard [1]. building forms evaluated by model tests”. Civil
Engineering (ASCE), May 1952, pp 42-46.
Based on a typical 3-pin frame system used in 3). M. Kanda and E Maruta, “ Characteristics of
storage sheds, the knee and centre rafter bending fluctuating wind pressure on long low-rise
moments (MK and MC) and horizontal (H) and buildings with gable roofs”, Journal of Wind
vertical (V) reactions at the base of the second Engineering & Industrial Aerodynamics, Vol. 50,
frame from the gable-end of building 173-182, (1993).
configuration No 1 are analysed here. The 4). J. D. Holmes, “Wind loads on the frames of
bending moments and horizontal and vertical large bulk-storage sheds”, 7th Australasian Wind
reactions are non-dimensionalised as Engineering Society Workshop, Auckland NZ,
2
C = M ( 1 U 2d w), C = H ( 12 U h 2h f w) October, 1998.
M 2 h H 5). J. D. Ginger, C. Hutton and G. F. Reardon
2
and C = V ( 2 U h dw) respectively, where w is
1 “Wind loads on large buildings with steep roofs”,
V 8th Australasian Wind Engineering Society
the width of the tributary area and hf is the height Workshop, Perth, February, 2000.
of the frame. Table 2 shows peak wind load 6). Standards Australia. Australian Standard SAA
effects for  = 45o and  = 90o, derived from the Loading Code Part 2 Wind Loads AS1170.2 -
“covariance integration” method [8] and 1989.
compared with that derived from AS/NZS 1170.2- 7). J. D. Ginger, “Wind loads on large storage
2001. Generally good agreement is seen, buildings”, CSTS Research Report, (2001)
although some underestimation by the 2001 8). J. D. Holmes and R. J. Best, “An Approach to
edition of the Standard is evident for the aspect the determination of Wind Load Effects on Low-
ratio of 6. Rise Buildings”, Journal of Wind Engineering &
Industrial Aerodynamics, Vol. 7, 273-287, (1981)
Table 1 Test building configurations and specifications

Config. Roof Span Mid roof Total Length (b), m Aspect Ratio
No. Pitch (o) (d), m height, m height, m (b/d)
1-Fig 1 35 40 22 29 96, 160, 240 2.4, 4, 6
2-Fig 2 50 50 15 29.8 150, 200, 250, 300, 350, 400 3, 4, 5, 6, 7, 8
3-Fig 3 Curved 45 22.1 29.1 108, 225, 360 2.4, 5, 8
4-Fig 4 15 80 20.4 25.7 108, 240, 350 1.35, 3, 4.35

Table 2 Wind load effects on Building configuration No. 1 Frame B versus aspect ratio

Load Effect AR Wind tunnel. AS/NZS1170. Load Effect AR Wind tunnel. AS/NZS1170.
Coeff. 2-2001 Coeff. 2-2001
 = 45o  = 0o  = 45o  = 0o
CMK 2.4 0.094/-0.062 0.126/-0.106 CH 2.4 -1.10 -1.55
4 0.121/-0.073 4 -1.36
6 0.139/-0.081 6 -1.52
CMC 2.4 0.072/-0.089 0.111/-0.128
4 0.082/-0.123  = 90o  = 90o
6 0.089/-0.146 CV 6 -0.92 -1.23

Fig 1. (left) Building Configuration No. 1, span =

40m, height = 29m, AR= 2.4, 4 and 6,  = 35o

6.0m

B
Fig 2. (right) Building Configuration No. 2, span
= 50m, height = 30m, AR= 3, 4, 5, 6, 7 and 8,
 = 50o

7.6m
 B
B

6.0m 6.0m

Fig 3. Building Configuration No. 3, span = 45m, Fig 4. Building Configuration No. 4, span = 80m,
height = 29m, AR= 2.4, 5 and 8, Curved roof height = 20m, AR= 1.35, 3 and 4.35,  = 15o

3.0 3.0

2.0 2.0

1.0 1.0

0.0 0.0
Cpeak
Cpeak

-1.0 0 15.95 31.9 47.85 63.8 -1.0 0 9.75 19.5 29.25 39 48.75 58.5 68.25 78

-2.0 -2.0

-3.0 -3.0

-4.0 -4.0

-5.0 -5.0

Distance along frame, (m) Distance along frame, (m)

AR=2.4 AR=4 AR=6 AR=3 AR=4 AR=5 AR=6 AR=7 AR=8

Fig 5. Peak positive and peak negative Cps on Fig 6. Peak positive and peak negative Cps on
windward and leeward halves of Frame B on windward and leeward halves of Frame B on
Building Configuration No. 1,  = 45o Building Configuration No. 2,  = 45o

3.0 3.0

2.0 2.0

1.0 1.0

0.0 0.0
Cpeak
Cpeak

-1.0 0 21.5 43 64.5 86 -1.0 0 28.2 56.4 84.6 112.8

-2.0 -2.0

-3.0 -3.0
-4.0
-4.0
-5.0
-5.0
Distance along frame, (m)
Distance along frame, (m)
AR=1.35 AR=3 AR=4.35
AR=2.4 AR=5 AR=8

Fig 7. Peak positive and peak negative Cps on Fig 8. Peak positive and peak negative Cps on
windward and leeward halves of Frame B on windward and leeward halves of Frame B on
Building Configuration No. 3,  = 45o Building Configuration No. 4,  = 45o