Windloadings and Hail
Windloadings and undertile membranes
The most important environment factor which affects the satisfactory performance of roofs is wind gusting.
During short-term wind gusts pressure differences are set up between the roof space (loft) and the outside of the roof covering. The result is a wind force that can cause the total or partial removal of the roof covering allowing further damage by natural elements. Where the roof pitch is below 30° this wind force results in suction on both the windward and leeward sides of the roof. This suction or lifting force, particularly on a low pitched roof, is often the most severe wind load experienced by any part of a building. Under strong wind gusts the uplift on the roof covering may be far in excess of the dead mass of these coverings, requiring both the roof covering and the total roof structure to be securely fixed to prevent the roof and/or covering from being lifted and torn from the building. Wind tunnel tests and practical evidence have shown that the satisfactory performance of a roof, and a tiled roof in particular, depends on the complementary function of the roof covering and the undertile membrane.
The results are discussed here briefly. In Figure A a roof with a pitch of less than 30° is experiencing a wind of velocity Vs metres per second horizontal and at right angles to the ridge line. The kinetic energy of the wind is transformed into a dynamic pressure q by the interaction of the roof obstruction with the moving wind: q (Newtons per m²) = P V²/2 where P = Density of air
The actual pressure experienced by any portion of the roof will vary with the roof pitch, height and plan dimensions of the building, and can be expressed as follows:
Pressure P = q.Cp where Cp = Pressure coefficient
The actual pressure coefficient Cp across the roof section will be the difference between the internal pressure coefficient Cpi and the external pressure coefficient Cpe.
Cp = (Cpe - Cpi)
If Cp is negative (Cpe less than Cpi) a suction or lifting force is experienced in the roof surface. The final expression for calculating the pressure being exerted on a roof structure under wind gusting is as follows:
P = q.(Cpe - Cpi). A where A = Area of surface
If a roofing undertile membrane is employed, the force on the roof covering will not be proportioned to
(Cpe - Cpi) but to (Cpe - Cpb) where Cpb = Batten cavity pressure coefficient.
The relative air permeability of the roof covering and the undertile membrane becomes the controlling factor. If the assumption is made that:
Permeability of roof covering = Permeability of undertile membrane, then Cpb will have a value midway between Cpe and Cpi. This will result in the lifting pressure on the roof covering being reduced by 50% based on this assumption. Wind tunnel measurements have shown that when a uniform pressure difference is applied across a concrete tile roof covering with an undertile membrane, almost all the load is absorbed by the undertile membrane. This is because the concrete tiles are more permeable to air than the undertile membrane. This results in the batten cavity pressure Cpb being very little different from the external pressure Cpe.
The presence of an undertile membrane therefore virtually eliminates the load on a concrete tile roof covering when experiencing an imposed uniform pressure difference between the internal roof space and the outside, provided that the undertile membrane does not give way. This protection is reduced by the presence of any hole in the undertile membrane.
Experience has also shown that under non-uniform and unsteady air velocities the response of the batten cavity pressure to the external pressure is very fast, with a relaxation time of around 0,02 sec. or less.
Wind tunnel measurements indicate that when an undertile membrane is employed with a roof covering the principal load on the roof covering is a result of the pressure differences generated by the interaction of the wind with the surface upstands.
When an undertile membrane is not employed with a roof covering, the normal internal pressure Cpi and external pressure Cpe relationship can in fact be reversed in certain situations. It has been found that the present South African roof construction that allows open verges and valleys in association with the normal abutment and eaves openings can result in air being preferentially removed via these dominant roof openings during gusting. Hence, the internal pressure Cpi will be less than the external pressure Cpe. Without an undertile membrane air will be drawn into the roof space via the gaps of the roof covering.
If it is raining at the time of this phenomenon, subject to gap sizes available, rain will follow the direction of the air into the roof space. In this situation the roof covering will be less susceptible to removal but rain penetration may occur.
A roofing undertile membrane (high tensile strength/tear resistance), performs a critical function in preventing roof coverings from being removed under high wind gusting and in some instances reduces the need for mechanical fixing. In areas of high driving rain, e.g. coastal regions, an undertile membrane will minimize the risk of rain penetration for all roof pitches that may occur as a result of the reversal of the internal/external pressure relationship caused by the other dominant roof openings. In order to withstand high wind loads it is necessary for all horizontal overlaps to be held down properly. One method is to use an additional batten over the overlap where necessary.
In addition, a suitable roofing undertile membrane will afford:
a) An increase in thermal insulation resulting in energy savings during winter and summer.
b) Reduced dust contamination in the loft space, hence allowing it to be utilised as a storage area.
c) Minimised water ingress and damage resulting from hailstones melting in valleys, concealed gutters, etc.
d) Protection against roof leaks in the event of damage to the roof covering.
The working performance of the roofing undertile membrane substantially reduces the lifting forces on the roof covering. In addition the undertile membrane brings definite advantages to the building. In essence an undertile membrane is an essential component of a pitched roof and should be considered an investment and an insurance for a weather-tight roof.
If a roof structure is fitted with an undertile membrane of suitable quality and is tiled according to the required specifications, it will withstand excessive wind speeds.
Hailstones
Hail, as a destructive force of nature, has plagued man, his crops and his property since the beginning of civilization. The vast majority of hailstorms contain hailstones that are relatively small. These small stones can damage crops, but not roofs.
It is known that thunderstorms and hailstorms are closely related and various meteorological phenomena related to thunderstorms and hailstorms e.g. dew point, cloud thickness temperature of cloud base, and temperature lapse rate, all reach maxima during the summer period. The maximum frequency occurs in the months of November and December when the temperature lapse rate and the surface temperatures are at their highest.
Figure A (Pretoria) indicates that hailstorms are almost entirely confined to the hours between midday and 22h00 with a maximum occurring around 17h00 to 18h00.
Figure B indicates the average number of hailstorm days per annum. It is clear that hailstorm frequency is closely related to height above sea level. Gauteng can expect 4-5 hail days per year whereas the coastal areas of KwaZulu-Natal can expect virtually none.
The ability of a hailstone to cause damage is directly proportional to its energy on impact and this in turn increases with the diameter of the hailstone. In brief a large hailstone is potentially a greater hazard than a small hailstone. The majority of hailstones studied have a density of 910kg m-³ indicating virtually clear ice. The shapes of hailstones are varied and although these must have a limited effect on the damage potential it is negligible compared with the overall effect of the hailstone diameter, i.e. terminal velocity versus impact energy.
Figure C indicates the relationship between hailstone diameter, terminal velocity and impact energy. These calculations assume a spherical model. Independent hail impact tests conducted by the SABS have indicated that a hailstone diameter of between 40-50mm and larger is necessary to damage standard Coverland concrete roof tiles.
Hailstorm statistics show that only 3% of all reports indicate hailstone diameters in excess of 30mm and only 0,6% indicate hailstone diameters in excess of 45mm. It is noted that these figures probably reflect upper limits as there is a natural tendency to ignore very light hailstorms.
The risk of a hailstorm containing hailstones of 45mm or larger, i.e. the critical size that is resisted by Coverland concrete roof tiles, is less than 6 in 1 000 hailstorms. Based on the hailstorm frequency of five per year in the highveld/bushveld regions, the risk is reduced to a chance of 1 in 33 years.
Hailstorms that tend to be very localized phenomena only become significant, so far as building roofs are concerned, when they occur in townships. The aforementioned risk, is further reduced by the chance of the critical one in thirty three year hailstorm falling in a township or the open veld.
The most densely populated area of the Highveld is Gauteng. If the land utilisation for residential purposes within this region is projected at 60%, the risk becomes once in 55 years.
Therefore the risk of a roof covered with Coverland concrete roof tiles being damaged by hail in the Highveld (high risk region) is less than once in a fifty year period, i.e. a period in excess of the nominal design life of a private dwelling.