STRUCTURAL - CONSTRUCTION TECHNOLOGY
DETERMINATION OF WIND LOADS ON POROUS SUNSHADE ROOF
COVER SHEETS USING CONVERSION FACTOR
VU THANH TRUNGa*
a
Vietnam Institute for Building Science and Technology
*
Corresponding author: email: trungvuthanh1975@gmail.com
Article history: Received 15/2/2023, Revised 22/3/2023, Accepted 26/3/2023
https://doi.org/10.59382/j-ibst.2023.en.vol1-1
Abstract: Porous sunshade roof cover sheets
installed on building roofs are new structures for
mitigating heat absorption. Thus, there have been
few studies on wind flow and its effects on wind-
permeable structures, and there is no provision for
these structures in the major international wind load
standards and codes. Unlike wind loading on
building roofs, wind loading on porous sunshade
roof cover sheets is a combination of wind
pressures on both the upper and lower surfaces,
which complicates the calculation of wind loading.
From the viewpoint of structural design, we need a
simple method for determining wind loading on
these structures. Wind tunnel tests have been
carried out for various wind angles, three porosities
(ratio between area of orifices and area of sheet),
and five gaps between sheet and roof, and results
have been analyzed. As a result, the author
proposes a simple formula for calculating wind
loading on porous sunshade roof cover sheets from
wind loadings on normal building roofs available
from wind loading codes and standards. The
calculated results based on this formula and the
Japanese wind loading standard (AIJ-RFLB
(2015))[4] are compared with those of wind tunnel
tests. There is a good similarity between them.
Keywords: conversion factor, wind loading,
porous sunshade roof cover sheet, low-rise building,
wind tunnel test
1. Introduction
Thermal reduction is always a problem for
building roofing systems, especially profiled steel
sheet systems. It reduces requirements for heating
and cooling system capacities, increases occupant
comfort, and even eliminates condensation on roof
surfaces in cold climates. Up to now, there are
several solutions for roof insulation such as loose
laid paver system, heat insulation roof tile and so
forth. But the cost of these solutions is still
Journal of Building Science and Technology - No.1/2023
expensive. Porous sunshade roof cover sheets have
been introduced to be a new solution. Fig. 1 shows
some pictures of application of the porous sunshade
roof cover sheets for a real roof of low-rise building.
There was some research on wind loads on porous
roofs or wind-permeable facades [1-3]. Wind loading
on this type of structure depends on the difference
between the pressures on the upper and lower
surfaces. It is therefore important to be able to
assess the influences of porosity and underneath
volume ratio on wind loading on these sheets. This
paper proposes a new formula to determine wind
loading on porous sunshade roof cover sheets.
2. Test set-up
A low-rise building model (200 mm high (H) 
470 mm wide (B)  710 mm deep (D)) with porous
sunshade roof cover sheets was tested in a
Boundary Layer Wind Tunnel, located at Tokyo
Polytechnic University, Japan. The length and
velocity scales were 1/50 and 1/4, respectively.
Terrain category III (power law index of 0.2) in [4]
was chosen for the tests. The turbulence intensity at
height 200 mm (10 m in full scale) was 0.26 and the
wind speed was 7m/s. Fig. 2 shows mean wind
speed and turbulence intensity profiles. There were
3 test model cases to consider the effect of porosities
(ratio between area of orifices and area of the sheet) of
the porous sunshade roof cover sheets (0%, 5% and
10%) with the gap between the sheet and the profiled
roof a = 4.7 mm (corresponding to underneath
volume ratios V* = a/(a+H) = 0.018) (see Fig. 3). And
there were 5 test model cases (porosity  = 5%) to
consider the effect of underneath volume. The
underneath volume was modified by adjusting gap
between the sheet and the folded roof, a, (see Fig.
3(d)). The values of a = 4.7, 5.7, 6.7, 8.7 and 10.7
mm for the folded roof corresponding to gaps
between the sheet respectively. They created
underneath volume ratios V* = a/(a+H) = 0.018,
3
STRUCTURAL - CONSTRUCTION TECHNOLOGY

0.023, 0.027, 0.037 and 0.046 respectively. The B, C and D) (see Fig. 3).
model had sixteen porous sunshade roof cover All test model cases were tested with a total of
sheets, one with 128 holes, while four porous 41 wind angles (0o to 360o in 10o steps and four
sunshade roof cover sheets had pressure taps (A, wind angles: 45o, 135o, 225o and 315o).

Profiled steel
sheet

Porous
sunshade roof Fixing
cover sheet

Hole

(a) A real arrangement of porous sunshade roof cover

(b) Close-up view of a porous sunshade roof cover sheet
sheets on a low-rise building
Fig. 1. Pictures of application of porous sunshade roof cover sheet

0.08 0.08
Experiment
Test (1/50) Experiment
Test (1/50)

Height above wind tunnel floor z (mm)

Height above wind tunnel floor z (mm)
AIJ-RFLB AIJ-RFLB
2004m U(z)= 1.7(z/450) 0.2U H
0.06 zg = 450 0.06 2004
(full scale)
U(z)= 1.7(z/z G ) aU H
(AIJ-RFLB(2015)
AIJ-RFLB 2004)
z/zg
z/zg

0.04 0.04
Roof height H = 10 m Roof
U(z)/U H
height H = 10 m Turbulence intens
(full scale) (full scale)
0.02 0.02

0 0
0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 0 0.1 0.2 0.3 0.4

Turbulence intensity:  z  Uz' / Uz

(a) Mean wind speed profile (b) Turbulence intensity profile
Fig. 2. Simulated wind conditions in wind tunnel

Porous sunshade Hole

roof cover sheet
B 0o
H = 200

D
C
A 

Wind
B = 470

D = 710 Building

(a) Geometry of test model (b) Close-up view of a porous sunshade roof cover sheet

4 Journal of Building Science and Technology - No.1/2023

 STRUCTURAL - CONSTRUCTION TECHNOLOGY

Y
CL

20
100
Hole Pressure tap Porous sunshade

10

1~7
a = 4.7 ~ 10.7
roof cover sheet Cpu

100
CL 0 o

10 470 10
Cpl


100

3.7
D B Pressure
tap 6 4 6
Wind

100
C A

20
160 160 160 160
X
20 10 10 710 10 20 (d) Detail of roof section
(c) Arrangement plan of porous sunshade roof cover
sheets
Fig. 3. Test model (all dimensions in mm)

Fig. 4. shows some pictures of a test model: test model in wind tunnel and a close-up view of a porous
sunshade roof cover sheet.

(b) Close-up view of a porous sunshade roof cover sheet

(a) Test model in wind tunnel
model
Fig. 4. Pictures of test model

3. Results and discussion The local reduction factor  L for different

porosities  was introduced to provide a good view
3.1 Local wind force coefficient
of reduction rate when comparing wind loading on
The local wind force coefficient on the sheet due the porous sunshade roof cover sheet with that on

to the combined effect of the upper and lower the building roof. The local reduction factor  L is
surface pressures is defined by:

Cn(i,t) = Cpu (i,t) - Cpl(i,t) (1) Cn
L 


(2)
where Cpu(i,t) and Cpl(i,t) are wind pressure C pu
coefficients at measurement tap i at time t on the 

upper and lower surfaces of the sheet, respectively; where C n is minimum peak local wind force

and Cn(i,t) is the local wind force coefficient at coefficient; and C pu is minimum peak upper surface
measurement tap i at time t of the sheet. wind pressure coefficient.

The simultaneous local reduction factor  L
The local wind force coefficients were defined as
analyses for all pressure taps for different wind
positive in the vertically downward direction.
angles are shown for porosities 0%, 5% and 10%
3.2 Local reduction factor respectively from Fig. 5.

Journal of Building Science and Technology - No.1/2023 5

STRUCTURAL - CONSTRUCTION TECHNOLOGY

The ensemble average values of local reduction 0.023, 0.027, 0.037 and 0.046, respectively.

factor  L are marked as () and the upper and Furthermore, values of standard deviation of  L
V

lower values are the standard deviation () from were from 0.05 to 0.14, from 0.05 to 0.15, from 0.05
the ensemble average values. For porosity  = 0%, to 0.15, from 0.07 to 0.20 and from 0.07 to 0.21 for
Fig. 5 shows that there is a gradual fluctuation in underneath volume ratios V* = 0.018, 0.023, 0.027,
ensemble average values from 0.34 to 0.54.
0.037 and 0.046, respectively. It means that
Similarly, ensemble average values varied from 0.27
ensemble average values and their standard
to 0.43 and from 0.22 to 0.38 for porosities 5% and
deviations of  L increase with increasing
V
10%, respectively. Furthermore, values of standard
 underneath volume ratio V*.
deviation of  L of porosity  = 0% were 0.08~0.16.
These values were 0.05~0.15 and 0.04~0.14 for From results shown in Fig. 5, we can get values

porosities 5% and 10%. It means that ensemble of  L with probability of exceeding 2% to be 0.75,

average values and their standard deviations of  L 0.62 and 0.55 for 0%, 5% and 10%, respectively
decrease with increasing porosity. (see Fig. 7). It indicates that local reduction factor
The local reduction factor  L for different
V
 L with probability of exceeding 2% decrease with
underneath volume ratios V* and porosity  = 5% is increasing porosity.

defined as same as  L in Eq. 2.
From the results shown in Fig. 6, we can get
For underneath volume ratios V* = 0.018, Fig. 6
values of  L with probability of exceeding 2% to be
V

shows that there is a gradual fluctuation in

0.65, 0.66, 0.69, 0.75 and 0.78 underneath volume
ensemble average values from 0.27 to 0.43. ratios V* = 0.018, 0.023, 0.027, 0.037 and 0.046,
Similarly, ensemble average values varied from 0.28
respectively (see Fig. 8). It indicates that local
to 0.44, from 0.29 to 0.47, from 0.31 to 0.51 and reduction factor  L with probability of exceeding 2%
V

from 0.33 to 0.52 for underneath volume ratios V* =

increase with increasing underneath volume ratio.

0.8 0.8
 L

Porosity  = 0%
 L

0.7
0
0.7 Porosity  = 5% 0
10
Local reduction factor
Local reduction factor

10
0.6 20 0.6

30 20
0.5 40 0.5

30
45
0.4 0.4
50

L
L

40
60
0.3 0.3
70 45

0.2 80 0.2
50
90 C n / C pu
C n / C pu

0.1 0.1


100


60



110
0 0
70
0 90 180 270 360 120 0 90 180 270 360

Wind angle  (deg) 130

Wind angle  (deg) 80
135
140 90
0.8
 L

150
100
160 0

Porosity  = 10%
0.7
Local reduction factor

170 110
10
0.6 180
120
190 20
0.5
200 130
30
210
0.4
135
L

220 40
225
0.3 140
230 45

0.2 240 150

50
250
160
C n / C pu

0.1


260 60


270 170
0
70
0 90 280
180 270 360
180
290
Wind angle  (deg) 80
300 190
90

L
310
Fig. 5. Variations of local reduction factor with
315 wind angle  for different porosities  200
100
320
210
330 110
6 Journal of Building
340 Science and Technology - No.1/2023 220
120
350
225
360 130
mean 230
135
STRUCTURAL - CONSTRUCTION TECHNOLOGY

 LV
0.8
0.8

 LV
0 0
*
0.7
V = 0.018 0.7
V* = 0.023

Local reduction factor

Local reduction factor
10 10
0.6
0.6
20 20
0.5
0.5
30 30
0.4
0.4

L

L
40 40
0.3
0.3
45 45
0.2
0.2
50 50

C n / C pu
0.1

C n / C pu

60 0.1


60


0
70 0
70
0 90 180 270 360
0 90 180 270 360

Wind angle  (deg) 80 Wind angle  (deg) 80

 LV

90

 LV
0.8 0.8 90

*
*
V = 0.027
100 0
V = 0.037 100 0
Local reduction factor

0.7 0.7

Local reduction factor

110 10 110 10
0.6 0.6
120 120
20 20
0.5 0.5
130 130
30 30
0.4 135 0.4
135

L

L
40 40
0.3 140 0.3 140
45 45
150 150
0.2 0.2
50 50
160 160
C n / C pu

C n / C pu
0.1 0.1



60 60
170



170
0 0
70 70
0 90 180 270 360 180 0 90 180 270 360
180

Wind angle  (deg) 190

80
Wind angle  (deg) 190
80

200 90 90
200
 LV

0.8

210 100 100

0.7
V* = 0.046 0 210
Local reduction factor

220 110 220 110

10
0.6
225 225
120 120
20
0.5
230 230
130 130
30
0.4 240 240
135 135
L

40
250 250
0.3
140 140
45
260 260
0.2
150 150
50
270 270
C n / C pu

0.1 160 160



60
280 280


0 170 170
290 70
0 90 180 270 360 290
Wind 180  (deg)
angle 180
300 80 300

 LV with wind angle  for different underneath volume

310 190 190
Fig. 6. Variations of local reduction factor 90
ratios V* for folded roof310
315 200 315 200
100

3.3 Conversion factor 320 210
110 320 210
where C n is minimum peak local wind force 330
220 330 220
Porosity and underneath volume ratio (i.e. gap coefficient for a porous sunshade120
roof cover sheet;
340
between sheet and building roof) are two important 225  130
340 225

factors to be considered when determining wind 230and C r is minimum peak 135 wind pressure coefficient for
350 350
230

loading on porous sunshade roof cover sheets. 240

360 a normal building roof. 360
240
140
mea
From the viewpoint of structural design, we need n a 250 From Eq. 3, minimum 150
peak local wind force mea
n 250
Mea
simple way (i.e. a simple equation) to determine the
n-
260
coefficient for a porous sunshade roof cover sheet Mea n-
is 260
std 160
wind loading on the sheet based on wind loading Mea
n+st
determined as follows. std
Mea
n+st
d 270 270
170 d
determined from [4] and porosity and underneath  

volume ratio of the sheet. Here, conversion factor

280
Cn   Cr 180
(4)
280

for the sheet with porosity  and underneath 290

Minimum peak wind pressure coefficient for a
190
290

volume ratio V* is defined as follows 300  200

300

310
normal building roof ( C r ) can be determined from [4], 310
 210
namely negative peak external pressure coefficient.
Cn

315 315
220
(3) The local wind force coefficients for porous
 320 320
225
Cr 330
sunshade roof cover sheets are very similar to wind
330
230

340 340
240
Journal of Building Science and Technology - No.1/2023
350
7 350
250

360 360
260

mea mea
270
STRUCTURAL - CONSTRUCTION TECHNOLOGY

pressure coefficients on the normal building roof 

  where C pu is minimum peak upper surface wind
( C r  C pu ). It means that conversion factor can be pressure coefficient.
determined as follows:
Conversion factors (local reduction factors)  L

Cn for porosity  and an underneath volume ratio V* =
  
(5) 0.018 are given in Table 1 and Fig. 7.
C pu

Table 1. Conversion factors  L for different porosities  and an underneath volume ratio V* = 0.018
Porosity  0% 5% 10%

L 0.75 0.62 0.55

Based on data shown in Table 1 and Fig. 7, a Furthermore, conversion factors (local

function of  L for porous sunshade roof cover sheet reduction factors)
 V
L for different underneath
was fitted as follows. volume ratios and a porosity  =5% are given in

 L = 2 + 0.74 (6)

Table 2 and Fig. 8.

Table 2. Conversion factors  LV for different underneath volume ratios V* and a porosity  = 5%
Underneath volume ratio V* 0.018 0.023 0.027 0.037 0.046
 V
L 0.65 0.67 0.72 0.78 0.8

Based on data shown in Table 2 and Fig. 8, a Combining two equations (6 and 7), we have the
function of  for porous sunshade roof cover sheet following equation:
V
L
was fitted as follows.  = 2 + 6.4V* + 0.62 (8)
 = 5.8V* + 0.55
V
L (7) Conversion factor  determined from Eq. 8 can
As already mentioned earlier, Eq. 6 is used for be applied for the sheet with different porosities 
cases that have an underneath volume ratio V* = and underneath volume ratios V*. It should be noted
0.018 and different porosities , whereas Eq. 7 is that Eq. 8 is only valid for porosity  in the range of
used for cases have a porosity  = 0.05 and different 0% ~ 10% and underneath volume ratio V* in the
underneath volume ratios V*. Therefore, it is difficult
range of 0.018 ~ 0.046.
to apply these equations to other cases. It needs to
have an equation of conversion factor  depending Based on Table A6.18 (1) of [4], plan of calculation
on porosity  and underneath volume ratio V*. zone of roof is shown in Fig. 9.

0.9 0.9 90o

angle angle 90o
0.8 = -2 + 0.8
0.74 Panel A Panel A
 LV
 L

0.7 0.7
Panel A Panel A
Conversion factor

0.6 0.6
Conversion factor

0.5 0.5
 LV = 5.8V* + 0.55
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
(o) (o)
0% 5% 10% 0.01 0.02 0.03 0.04 0.05
Porosity 
Underneath volume ratio V*
L
Fig. 8. Variation of local reduction factor  L with
V
Fig. 7. Variation of local reduction factor with probability
of exceeding 2% for different porosities  and an probability of exceeding 2% for different underneath
underneath volume ratio V* = 0.018 volume ratios V* and a porosity  = 5%

8 Journal of Building Science and Technology - No.1/2023

STRUCTURAL - CONSTRUCTION TECHNOLOGY

Tables 3, 4 and 5 show comparisons of  AIJ

minimum peak local wind force coefficients for Cn and from wind tunnel test for normal
porous sunshade roof cover sheets obtained  WT
from wind tunnel test, from [4] (using building roof (using conversion factor ) C n for
conversion factor  ) and from wind tunnel test all zones were higher than absolute values of
for normal building roof (using conversion factor 
 ). It is clear from this table that all absolute C n from wind tunnel test. It means that Eq. 8
values of minimum peak local wind force can be applied to determine wind loading on
coefficient from [4] (using conversion factor ) porous sunshade roof cover sheet.

Fig. 9. Plan of calculation zone of roof (all dimensions in mm)

Table 3. Comparisons of minimum peak local wind force coefficients for porous sunshade roof cover sheets for normal building
roof for corner zone
Porosity  0% 5% 10% 5% 5% 5% 5%
Underneath volume ratios V* 0.018 0.018 0.018 0.023 0.027 0.037 0.046
Conversion factor
0.74 0.64 0.54 0.67 0.69 0.76 0.81
 = 2 + 6.4V* + 0.62
Minimum peak wind pressure coefficient on the
 AIJ 5.4
normal building roof Cp (from [4])
Minimum peak wind pressure coefficient on the
 WT
5.9
normal building roof C p (from wind tunnel
test )
Minimum peak local wind force coefficient (from [4] and )
 AIJ  AIJ 4 3.4 2.9 3.6 3.7 4.1 4.4
Cn = C p
Minimum peak local wind force coefficient (from
 WT  WT 4.3 3.7 3.2 3.9 4.1 4.5 4.8
wind tunnel test and) Cn = Cp
Minimum peak local wind force coefficient from
 2.4 2.4 1.9 2.5 2.7 2.9 3.7
wind tunnel test Cn
Table 4. Comparisons of minimum peak local wind force coefficients for porous sunshade roof cover sheets for normal building
roof for edge zone
Zone Edge zone
Porosity  0% 5% 10% 5% 5% 5% 5%
Underneath volume ratios V* 0.018 0.018 0.018 0.023 0.027 0.037 0.046
Conversion factor  = 2 + 6.4V*+ 0.62 0.74 0.64 0.54 0.67 0.69 0.76 0.81
Minimum peak wind pressure coefficient on 3.2

Journal of Building Science and Technology - No.1/2023 9