4.1.1 Structure types
Basically, two types of structure are used:
a) The beam system, where simple beams are stressed on bending. Both tensile and compressive stresses occur on the beam.
b) The system using trusses. In this case the structural members of the truss receive either tensile or compressive stress only.
Structure types; Principles of trusses
4.1.2 Principles of trusses
The basic shape of the truss is triangular, because a triangle cannot be distorted, unlike shapes with more than three angles.
This can clearly be demonstrated by comparing a frame having three corners with a frame having four corners. The triangle frame is always rigid. The rectangular frame can easily be distorted, the only resistance being the corner joints.
Perfect and imperfect structures
Structures consisting of triangles only are called perfect structures. If a structure includes elements of more than three angles, it is called an imperfect structure.
There are many examples of perfect structures, e.g. bicycle frames or scaffold braces.
Trusses should be made as perfect structures, consisting of triangle-shaped elements only; rectangular or square structures should be supplemented with diagonals.
Perfect and imperfect structures
4.1.3 Structural design in a step-by-step approach
When designing the roof structure, the supporting elements required by the roof-cover material must be taken into account.
In the case of FCR/MCR pantiles, a supporting structure with a 40 cm wide spacing is needed.
Structural design in a step-by-step approach
Since the upright structural parts of any building (walls, columns) have a larger span, a roof structure is required to span horizontally. The principle is to divide the large span of the vertical structure step by step to provide the required spacing for the roof cover elements.
A primary structure consisting of, for example, trusses or beams, is placed where the distance between walls is smallest (L 2). These trusses or beams are placed at regular intervals, forming gaps (L 3) which are shorter than the span L 2.
A secondary, finer structure (e.g. ridge beams and purlins) is
laid dividing the gaps further (L 4).
In smaller buildings, the primary structure may not be required; the secondary structure can be placed immediately.
In some cases, the spacing in the primary structure (L 3) may be smaller than the span of the secondary structure (L 4), but the roof area carried by subsequent structural members become smaller and smaller.
The system of dividing the span further can be continued by placing. rafters with the spacing L 5.
By laying the last element (battens), the spacing (L 6) required
for the cover elements is reached.
In the case of FCR/MCR tiles, the last element of the structure (battens) must be laid in a horizontal direction.
4.1.4 Detail principles
First large, then small members
The above-described principle of reducing the span step-by-step has the following consequences:
The primary structure carries the highest load and is hence the
strongest, i.e. heaviest member.
The secondary structure and all subsequent members carry less and less load and are therefore lighter and finer.
Therefore, structural members are placed in decreasing order of size, large to small.
Laying of rectangular beams
Beams of a rectangular section are laid with the larger dimension in the vertical direction and the smaller dimension in the horizontal direction. In this way the load bearing capacity is greater.
First large, then small members; Laying of rectangular beams
The function of bracing is to prevent horizontal movements in the roof structure caused by winds, earthquake or other horizontal forces.
The principle of bracing is usually based on triangulation. Braces can be laid either in a vertical position between trusses (a) or in an inclined position along the rafters (b).
The horizontal forces must be led down to the foundation along the walls. Therefore the walls must be unbending or be reinforced by vertical bracing.
4.1.6 Special forms of primary structure
A single main truss
In some cases it may be economical to lay a main truss (girder) along the longer span of a rectangular shape. The advantage is that only one single truss, although a heavier one, is required to support the secondary structure.
Such trusses have to be designed by a qualified engineer.
A single main truss
For the construction of large buildings, halls etc., frame trusses can be used. The trusses rest directly on the foundations and provide support for the roof as well as for lightweight walling systems.
Spanning in two directions
To span a space which is square or almost square, it may be appropriate to lay the primary structure in two directions. In concrete technology this system is common, when beams or slab reinforcements are laid in two directions.
Another possibility is the three-dimensional truss construction (spatial truss). In this case the crossing points of the structural members require a careful design and workmanship, especially if they are made of timber.
Frame trusses; Spanning in two directions
(also see Chapter 2.3 and 3.2)
As described in Chapter 2.3, structures have to withstand not only live loads and dead loads, but also loads from storms and earthquakes.
[see FCR News No 5] Winds with a speed of more than 75 km/h are considered storms. They cause heavy pressure and suction on the structure in a direction perpendicular to the surface.
The main storm-prone areas in the world are well known, such as Southeast Asia, Central America, where special care has to be taken.
Storms can, however, occur everywhere, although perhaps less frequently and with less intensity. Therefore structural means with which to avoid destruction of the roof are important in any location.
Structures are normally designed for a wind speed up to 150 km/h.
Means for avoiding storm damage:
The roof structure should be secured to the walls by bolts or rods against uplift (see also chapter 4.3.4). These bolts are attached to the ring beam reinforcement. Where there are no ring beams, the anchorage can be built into the masonry wall. In this case the anchorage should be at least three brick courses deep.
Possible example of wind load distribution; Insufficient anchorage of wall plate, in top course only
Based on the data in Chapter 2.3, the magnitude of the uplifting forces can be determined. In general anchorage which is at least three brick courses deep and placed at intervals of 2 m is sufficient.
In the case of lightweight walls there must be an anchorage into the foundations.
Winds naturally have a strong horizontal load component on roofs. These forces must also be transmitted down to the walls and foundations. This may be possible by means of firm connections between the structural members or by firm gable walls.
A more secure and therefore recommended method is the use of bracing.
Anchorage 60 cm deep is recommended; Lightweight wall with anchorage in foundation; Bracing in the plane of the roof slope
At the point where bracing is connected to the walls, concentrated forces occur. These forces must be transmitted along the walls to the foundations. This is possible either by firm walls or by additional bracing in the walls.
c) Ring beam
A ring beam (tie beam) is a continuous beam at the top of the wall.
It provides a solid base on which wall plates or rafters can be anchored. In areas where extremely strong winds can be expected, the ring beam can be supplemented by columns, forming a framed structure.
Such structures are usually made of reinforced concrete.
Bracing of trusses in a vertical plane; Framed structure
Influence of roof pitch
Wind causes suction on the lee-side of the roof with a danger of the roof lifting off. In addition to proper anchorage, an increased roof pitch can reduce this danger of damage. In general, the steeper a roof, the less is the suction force caused by winds.
Shape of the roof
The edges (ridge, eaves, verge) of the roof are more susceptible to wind than the surface. The most critical parts are the verge and the mono-pitched ridge. Therefore for wind safety the preferred roof shape is the hipped roof. Gable roofs or mono-pitched roofs are disadvantageous.
(also see Chapter 3)
Earthquakes occur mainly in well known zones around the globe. In these areas special precautions have to be taken.
Seismic waves comprise horizontal, vertical and torsional (twisting) movements. Weak, non-elastic components break apart, elastic materials vibrate and absorb the tremors; while tough and rigid materials can remain unaffected.
Means to avoid damages by earthquakes
a) Simple building form
Simple and symmetrical buildings are safer than buildings with complicated and asymmetric shape.
Buildings of complicated shape should be divided in simple independent components which are separated by expansion joints.
Simple building form: Good - Bad
b) Light roof
For safety reasons in earthquake areas, the centre of gravity (see Chapter 2.1.3) of the building should be as low as possible. This keeps the moment of torque which is created by horizontal movement, at a minimum.
The roof should therefore be light. FCR/MCR is an advantageous material in this respect. If the roof is properly anchored and braced for storms (see chapter 4.2.1), then it is also safe for earthquakes.
c) Flexible structure
A structure with flexible but tight joints can balance horizontal movements by giving way to them. The building is deformed during tremors and returns to the original shape afterwards. Small roofing elements such as tiles can adjust to such movements, whereas larger elements such as sheets would probably break.
Bamboo and pole timber are suitable materials for such building systems.
d) Stiff structure
Another approach is to build a structure with stiff joints to avoid any deformation during earthquakes. This results in dynamic forces which demand a firm construction.
Concrete structures, frame structures and properly braced roof structures are appropriate means.
e) Independent roof support
Alternatively, the roof structure can be fixed to independent supports which are structurally separate from the walls. In the case of the walls collapsing, the roof remains unaffected.
Stiff structure; Independent roof support
This chapter deals with sawn timber only. Another kind of timber, pole timber, is briefly dealt with in Chapter 4.4 Alternative structures.
Timber is not only one of the oldest, but also one of the most versatile building materials. However, it is an extremely complex material, available in a great number of varieties and forms with greatly differing properties.
Timber is basically a renewable material. Nevertheless, there is an universal concern about the rapid depletion of forests and the great environmental, climatic and economic disaster that would follow. Although construction timber represents only a small fraction of the timber felled, it should be used thriftily and wastage should be minimised. In addition reforestation programmes should be promoted.
The cross-section of a trunk or branch reveals a number of concentric rings. The trunk diameter increases by the addition of new rings, usually one ring a year.
The early wood (spring wood) formed during the main growth period has larger cells, while in the dry season the late wood (summer wood) grows more slowly, has thicker cell walls and smaller apertures, forming a narrower, denser and darker ring, which gives the tree structural strength.
Structure of a tree trunk (hard wood and soft wood)
As each ring forms a new band of active sapwood, starch is extracted from an inner sapwood ring, adding further to the inactive heartwood core. Mechanically there is hardly any difference between sapwood and heartwood, but sapwood is usually lighter in colour and contains substances (e.g. starch, sugar, water) which attract fungi and certain insects.
The slower the tree grows, the narrower are the growth rings, and the denser and stronger is the timber. Its resistance to biological hazards is also usually greater.
(Timber commonly used in roof construction see Appendix 2)
Types and properties of timber
Soft and hardwood
Woods are classified as either hardwood or softwood. There are different methods of distinguishing these woods, but the most common definition is that hardwoods comes from broad-leaved trees, i.e. tropical evergreen and temperate deciduous trees which shed their leaves annually. Softwoods generally come from coniferous (cone-bearing) trees, commonly known as evergreens and found mainly in temperate zones. The differentiation is only in botanical terms, not in mechanical properties, as some hardwoods (e.g. balsa) are much softer than most softwoods.
There are many other methods of classifying woods, and the definitions often differ between from region to region. For instance in West-Africa the terms red wood for hard and strong timber and white wood for soft and weak timber are common. In other regions this terminology is not known, or used in another sense.
Timber for building construction can be divided into two or more categories according to its mechanical strength. Often one distinguishes between primary and secondary timbers.
Primary timber generally comes from slow-growing, aesthetically appealing hardwoods which have considerable natural resistance to biological attack, moisture, movement and distortion. As a result, it is expensive and in short supply.
Secondary timber comes from mainly fast-grown species with low
natural durability. With appropriate seasoning and preservative treatment,
however, the timbers physical properties and durability can be greatly
improved. With rising costs and diminishing supplies of primary timbers, the
importance of secondary species is rapidly increasing.
When sizing structures, the bearing capacity of the timber being used has to be taken into account. This should be done by consulting the local norms and standards.
In remote areas, data regarding local timber are often not available. For such cases Chapter 4.3.4 provides a simple method for classifying timber without sophisticated equipment. The results may be used as a sizing aid to optimise structures of simple buildings. Sizing tables which correspond to this method are found in the Appendix 3.
Sawn timber products
Sawn timber is cut mainly from the trunks of older trees with large diameters, in rectangular sections as beams or boards. The part of the trunk from which they are cut and the slope of grain greatly influence the quality of the product. Therefore, when selecting timber, the direction of the grain has to be observed.
Quality of boards
High quality sawn timber is used for heavily-stressed structural members, e.g. purlins and trusses.
Low quality timber is used for wall plates etc.
Quality of beams
To make the best use of a trunk, the cutting method has to be considered. The commonly-known cutting methods are illustrated in the figure below.
Selection of timber
For structural members which are under high stress, such as purlins, rafters and trusses, the selection of suitable timber is of great importance.
Timber with cracks, knots or with grains that are not longitudinal should not be used. Such timber should only be used in positions with reduced stress, such as wall plates.
During felling and transport cracks may occur. Such timber should be rejected.
Cracks may also occur due to shrinkage which is unavoidable. Such timber should be tolerated to a certain extent, but not used for heavily-stressed structural parts.
Hidden cracks may also occur but are very difficult to detect. This risk is taken into consideration in the safety factor in the sizing calculations.
The strength of beams can be greatly reduced by knots, especially when located in the area of the greatest bending moment and in areas with tensile stress.
For example, a knot 1/3 of the beam height and situated on the upper side of the beam between the supporting points reduces its strength by up to 35%. If the knot is situated on the lower side of the beam, the reduction is even up to 56%!
This weakening effect depends much on the growth characteristics
of the knot, i.e. how well it is grown into the adjoining wood.
(also see example in Chapter 4.3.4)
Direction of grain
The strength of timber also depends on the direction of the grain. It should be longitudinal. If it is not, the strength is drastically reduced.
Direction of grain
Juvenile and cambium
Timber containing juvenile and cambium parts should be used for subordinate structural members only, e.g. wall plates.
To avoid cracking and warping, only seasoned timber should be used.
Wind thrown wood
For ecological reasons it would be highly sensible to use timber from trees blown over by wind. However, because of the risk of invisible cracks it should only be used for subordinate structures, rafters, battens etc. It should be rejected for use as heavily-stressed beams and trusses.
Such invisible cracks are difficult to detect. A simple but not very reliable method is to throw the cut beams from a height of 0.5 or 1 m. Badly-cracked timber will fall apart. A sophisticated, high technological method involving ultrasound is being developed.
4.3.2 Timber preservation and seasoning
Timber is a highly valuable and durable material. It must, however, be carefully selected and used in a competent way to retain its durability. The following aspects are important in these respects:
Timber which is cut in the non-growing time (winter) is more durable.
Cambium parts should not be used.
Timber should be properly seasoned before use.
Care should be given to structural details.
In certain cases chemical treatment of the timber may be necessary.
Before manufacturing timber components, the timber has to be properly seasoned. One reason is that during drying, timber shrinks. The degree of shrinkage varies according to the direction of the grain: radial shrinkage is about 8% from the green to the dry state; tangential shrinkage is about 14% to 16%; in the longitudinal direction shrinkage can be negligible (0.1% to 0.2%). The use of unseasoned timber results in cracks, warping and as a consequence uneven roofs.
If timber is to be chemically treated, this must be done only with seasoned timber. Some wood preservation methods do not allow the timber to dry properly. The moisture is trapped because the chemicals hermetically seal the timber surface. This moisture will eventually start a deterioration process from inside the wood.
By proper seasoning the moisture content of timber is reduced to its equilibrium moisture content (between 8% and 20% by weight, depending on the timber species and climatic conditions). This process, which takes from a few weeks to several months (depending on timber species and age, time of felling, climate, method of seasoning, etc.), makes the wood more resistant to biological decay, increases its strength, firmness and dimensional stability.
Basically there are three different methods for seasoning:
Air seasoning is done by stacking timber so that air can circulate around every piece. Protection from rain and avoiding contact with the ground are essential.
Forced air drying is principally the same as air seasoning, but the rate of drying is controlled by stacking the timber in an enclosed shed and using fans.
Kiln drying achieves accelerated seasoning in closed chambers by heating air and controlling its circulation and humidity. This reduces the seasoning time by 50% to 75%, but incurs higher costs. An economic alternative is to use solar-heated kilns.
The time requirement for seasoning is greatly reduced if the timber is felled in the non-growing season (winter), when the moisture content of the tree is low.
Seasoning methods; Structural means
b) Structural means
The most important aspect in preserving timber is to use it correctly in an appropriate building concept and detailing.
Protection from moisture
Access to circulating air
Avoiding of contact with ground
Protection from termites by concrete ring beams or galvanised iron sheets
Protection from moisture
Moisture occurs in the form of rain, humidity from walls and from dampness rising from the ground. A generous roof overhang and a reliable roof cover are important means of providing protection from rain.
Where timber parts come into contact with walls , they should be
separated by a bituminous felt.
Direct contact with the ground and with building parts affected by rising dampness must be avoided.
Access to circulating air
Where air cannot circulate or is excluded, a humid atmosphere forms. This promotes the development of fungus. Therefore timber should never be entirely enclosed with airtight materials and cavities should be ventilated.
Avoiding of contact with ground
The ground usually contains moisture which rises along any hygroscopic material such as timber, masonry etc. that is in direct contact with it. It also harbours different kinds of insects such as termites that can quickly destroy timber. Timber must therefore not come into contact with the ground.
c) Choice of material
Whenever available, timber should be used which is resistant to fungal and insect decay. Appendix 2 provides information in this respect.
d) Preservative treatment
When considering preservative treatment of timber, it should be remembered that timber is the healthiest of all building materials and it is paradoxical to poison it. This applies particularly when other methods could be used to protect it; for instance, a good building design (excluding moisture, good ventilation, accessibility for periodical checks and maintenance, avoiding contact with the ground, etc.) and a careful selection of the timber.
However, in the case of extreme climatic conditions and heavy fungal and insect attack, seasoning and structural means alone may not be sufficient to protect timber, particularly from species which are less resistant. Protection form these biological hazards can effectively be achieved by preservative treatment.
When preservative treatment is required, non-chemical and non-poisonous methods should be given priority.
Stacks of timber can be smoked above fire places or in special chambers, destroying the starch and making the surface unpalatable to insects. However, cracks can occur which eventually facilitate insect attack.
Immersion of timber in (preferably flowing) water for 4 to 12 weeks removes starch and sugar which attract borer beetles. Large stones are needed to keep the timber submerged.
Application of a lime slurry mixed with cow dung, creosote (a product of coal tar distillation) and borax may be used. This must not be done indoors, because of the strong odours.
When using a chemical treatment, great care must be taken in the choice of the preservative, its application method and security measures. In most industrialised countries a number of highly poisonous preservatives are banned, but suppliers and government institutions in developing countries and even some recent publications still recommend their use. No chemical preservative should be used without the full knowledge of its composition. Those containing DDT (dichlor-diphenyl-trichlorethane), PCP (pentachlorphenol), lindane (gamma-hexa-chloro-cyclohexane), arsenic, quicksilver, lead, fluorine and cadmium should be avoided.
Research on non-poisonous preservatives is still underway and full clarity on the toxicity of the recommended and currently available chemicals has not yet been attained. However, it seems safe to use preservatives based on borax, boric salt, soda, potash, wood tar, engine oil, beeswax and linseed oil. Their resistance to biological agents is less than that of the poisonous chemicals mentioned above, but can be equally effective in conjunction with a good building design.
There are several methods of applying chemical treatment to timber. Some examples are:
Immersing in a preservative solution
4.3.3 Timber structure types
The design of timber roof structures depends mainly on the span which is to be covered and on the properties of the timber. The procedures described in Chapter 4.1.3 can be analogously applied.
a) Battens only (up to 1 m or 1.5 m span between gable walls)
The simplest structure is spanning battens from gable wall to gable wall. This is possible for very small buildings, e.g. toilets or small annexes like porches.
Battens only; Rafter and batten
b) Rafter and batten (up to 4 m span between walls at eaves)
For small buildings with a span of up to approx. 4 m, the use of rafters placed directly from wall to wall is possible. This results in a mono-pitched roof.
The span between gable walls is arbitrary. The maximum spacing between rafters is 1 m - 1.5 m.
c) Purlin and rafter (up to 5 m span between gable walls or other supporting structure)
For medium-sized buildings purlins can be used to support the rafters. This allows a free span between gable walls of up to 5 m.
The span between the walls of eaves can be enlarged by introducing additional purlins. This, however, leads to very high ridges.
e) Truss, purlin and rafter
Large buildings can be roofed with conventional trussed structures. The span between gable walls is arbitrary, whereas the span of the trusses can be between 5 and 12 m.
Special truss design allows even larger spans. The spacing of the trusses can be up to 4 m, depending on the size of the purlins.
Purlin and rafter; Truss, purlin and rafter
f) Trussed rafters
lnstead of using a truss-purlin-rafter construction, each pair of rafters is tied together with a horizontal tie beam and acts similarly as a truss.
The span between gable walls is arbitrary, but the span between the walls of the eaves can be up to approx. 6 m. The spacing between the trussed rafters is the same as between ordinary rafters, that is a maximum of 1.5 m.
For larger spans more braces must be used.
Many different forms and systems of truss construction are possible, using various levels of technology (see specialised literature ).
g) Industrial timber elements
A number of systems are used which widen the scope of timber as a construction material. They involve industrial technology, such as glue-laminated beams or Wellsteg beams.
Industrial timber elements
4.3.4 Sizing of timber structures
In consultation with LIGNUM, Zrich, and P. Hsler, Bubikon, Switzerland
The sizing of timber structures is usually a complex engineering task and requires expertise. The main factors that have to be considered are
properties of the timber,
loads on the structure and
the structural system.
Situation in practice
All over the world, hundreds of different types and qualities of timber are used for construction, all with highly varying load-bearing capacity. The bearing capacity of the same kind of timber can also vary, depending on the climatic and soil conditions of the area where the tree has grown.
Often the properties of the available timber are not known to the user, and standards are often either not available or do not exist. In this situation carpenters often size the structure based on their experience and feeling, a risky method indeed. Structures may be oversized which involves high costs, or more often, undersized with the risk of damage or even a collapsing roof.
To overcome this dangerous situation this chapter provides a relatively simple method of sizing timber structures when the properties of the timber are not known.
The method consists of two steps:
The timber has to be tested and classified accordingly . The test can only be omitted if the timber species is unmistakably known and can be classified according to Appendix 2 (Timber commonly used in roof construction).
Once the timber has been classified and the structural system (span) is known, the correct size can be established with the help of the tables in Appendix 3.
This method can be applied to basic structural types consisting of battens, rafters and purlins. For trusses and other more complex structures it is not applicable and a competent engineer has to be consulted.
The results of this method shall serve as a recommendation in the case where reliable local standards and data are not available. The results are not to be considered as a binding standard, and neither the author nor the publisher can take responsibility for damages.
1. Step: Testing Using a simple test, the breaking load is established.
A piece of timber, 40 x 40 mm, is placed over supports which are at a distance of either 2 m or 3 m. In the centre a load is applied and increased until the timber breaks. This breaking load determines the classification.
Where laboratory facilities are not available, this test can be done with simple improvised methods. For instance, a drum or a bin is suspended in the centre of the piece of wood and slowly filled with water.
An example of an improvised test method is illustrated below.
A wheelbarrow is suspended from the timber being tested and filled with any kind of weight. The wheelbarrow and the weight are weighed by a spring balance.
Test: A timber beam is spanned over 2 m or 3 m, then a load is applied in the centre and increased until the timber breaks.
At least 6 tests have to be carried out with timber from different trees of the same species because the properties can vary greatly.
Test results with Swiss pine
To classify the tested timber, the relevant breaking load is calculated as follows:
Take the lowest load and add 20% of the load difference between the lowest and the average.
The formula to calculate the relevant load is thus:
P1 + 0.2 |f (P1+P2+P3+P4+P5+P6,6) - P1
Thus for the above example:
134+0.2 |f(134+140+148+148+168+180,6) -134 = 138 kg
Classification of timber according to test result
above 261 kg
above 174 kg
211 - 260 kg
141 - 173 kg
130 - 210 kg
87 - 140 kg
(For classification of commonly used timber species also see Appendix 3)
Conditions of test:
The timber being tested must be free of knots at least in the area near the centre where the load is applied.
Timber with knots not to be used for test; Timber with fibres not longitudinal must not be used in the test
The fibres of the timber being tested must be in the longitudinal direction only.
The test timber must be free from fungi and insect
The test timber must be free from cracks.
The humidity of the test timber must be the same as that of the timber used for construction. Extra seasoning of the test timber would result in values which are too high.
The load must be applied within 5 minutes. A long testing period would give false results.
Whenever a new, unknown type of timber with questionable properties is used, the test should be repeated.
2. Step: Sizing When the timber has been tested and classified, sizing with the help of the tables in Appendix 3 can be done. Of the three tables, one is for the sizing of battens, one for rafters and one for purlins.
How to read the tables
To read the tables, the free span (L) of the beam has to be determined. Then the spacing (c) between the beams (centre to centre) has to be defined. Entering these two values and the respective timber category in the table gives the required timber dimensions.
If the dimensions of the available timber are smaller than required, the span and/or the spacing have to be reduced, and vice versa.
Basis of the tables in Appendix 3
The calculation of the tables is based on the loads listed below (also see Chapter 2.3.3). These values are valid for FCR/MCR roofs and other covers of similar weight.
40 kg/m2 (0.4 kN/m2)
80 kg (0.8 kN) in the centre or at the end of a cantilever
50 kg/m2 (0.5 kN/m2)
30 kg/m2 (0.3 kN/m2)
50 kg/m2 (0.5 kN/m2)
80 kg (0.8 kN) in the centre or at the end of a cantilever whichever creates the bigger bending moment
55 kg/m2 (0.55 kN/m2)
30 kg/m2 (0.3 kN/m2)
Dead, live load
55 kg/m2 (0.55 kN/m2)
80 kg (0.8 kN) in the centre or at the end of a cantilever whichever creates the bigger bending moment.
Rafters and purlins
A wind of 150 km/hr can cause a wind pressure of 30
kg/m2 in a closed buildings. In completely open buildings the
pressure can be much higher. Suction can reach up to 70 kg/m2, but it
is not taken into consideration because the resultant is reduced by the dead
load. However, appropriate anchoring of the roof must be provided.
(Also see Chapter 2.3.2)
Snow load and extreme wind load (higher than 150 km/h) are not accounted for in the table.
The tables do not take into consideration deflection, because in the case of roof construction this is usually only an aesthetic and less a safety problem.
The tables are based on the following static system:
Variable number of supports with the span L, cantilever maximum 0.30 L.
2 or 3 supports with the span L
Cantilever (Lc) maximum 0.35 L.
In the case of more than 3 supports the effective span (Leff).can be reduced by 15% when reading the table.
2 or 3 supports with the span L
Cantilever (Lc) maximum 0.40 L.
In the case of more than 3 supports the effective span (Leff).can be reduced by 15% when reading the table.
Rafters and purlins; Roof slope
A roof slope between 25o and 40o is considered.
The tables in Appendix 3 provide timber sizes that take into account the natural irregularities of timber, such as varying density, humidity, direction of growth rings and fibres, and irregularities in the structure of the timber, by applying an appropriate security factor.
The smaller the safety factor that is applied, the less timber is needed and the lower are the construction cost; the risk of damage, however, is greater.
The tables therefore provide two levels of security:
In general a normal security factor is used. These values are printed in bold in the table. These values should normally be used.
In subordinate structures a reduced security factor may be used. These values are shaded in the table. If these values are applied, savings can be made in the timber used. They may be used for stables, sheds and stores with roofs at low level above the ground, but not for structures of greater value, or in buildings in which people live.
The safety factors do not take cracks and large knots into account. They can greatly reduce the strength of the timber. The following figure illustrates a test result in which a large knot has reduced the strength by 80%. Compare this to well selected timber as shown before under 1. step: Testing.
The use of such faulty timber is highly dangerous for the roof structure and even for the life of the workers.
Test with faulty timber and the result: The strength is only 20% of that of well selected timber.
4.3.5 Timber structure details
This chapter describes common detailing for the basic elements of simple timber structures. The main elements are:
Wall plates - these are fixed on to the wall to receive the lower end of the rafters.
Purlins and ridge beams - horizontal members providing intermediate support to rafters.
Rafters - similar to beams but inclined.
Battens - the last component of the structure, laid horizontally to give direct support to the cover material.
Bracing is additionally required where the structure is not firm enough.
For larger spans additional trusses are required to support the roof.
Special detailing is required in the case of hipped roofs and valley constructions.
Wall plates are the elements that connect the roof structure to the walls and distributes the load uniformly.
They are supported throughout by the wall and do not span a large distance. If there is no supporting wall but only columns, it is called purlin.
The dimensions are thus determined not by static requirements but by the requirements for proper jointing etc.
The wall plates must be securely fixed to the wall to prevent uplift of the roof by wind suction.
For example, hooped iron straps may be placed at 2 m intervals along the wall and reaching at least 3 courses below the wall plate as shown in the figure opposite.
Simple lengthening joints can be used for wall plates. Possibilities are:
a) Halving joint
b) Common scarf
c) Stopped scarf
Wall plates as tie beam
For smaller buildings, wall plates may also serve as a tie beam when a concrete tie beam is not required. In this case the joints have to be firm and built in a way that horizontal stress can be borne.
a) Hook halving joint
b) Stopped hook scarf
The wall plate, if it functions as a tie beam, must run all around the building and have strong corner joints (c). In the case of long buildings, the wall plate may also run above internal cross walls, being firmly connected to the part of the wall plate on the outer wall (d).
Wall plates as tie beam
Joints can be strengthened by fish plates; enabling them to take care also for shear stress.
a) One-nailed fish plate
This has limited strength and results in an asymmetrical distribution of the shear load on the beam.
b) Two-nailed fish plate
This has about double the strength compared to the one-nailed fish plate and is also used in lengthening other components of the structure such as purlins and rafters.
Fish joints and rules
Rules for fish joints
a) The length of the fish plates should be at least 5 times the height of the joint member.
b) The nails should be as long as to enter 3/4 of the thickness of the joint member and 12 times the diameter () of the nail.
c) Nails should be evenly distributed over the entire fish plate, normally in a staggered manner.
d) The spacing between the nails, and between the nails and the edge of the fish plate depends on the diameter () of the nails:
Along the grain of the timber, the distance between the nails should be 10 times the diameter () of the nail.
The distance between the nail and the edge of the fish plate
along the grain 10 times the diameter of the nail,
across the grain 5 times the diameter of the nail.
Purlins and ridge beams
Purlins and ridge beams form the primary structure (also see Chapter 4.1) of the roof, spanning horizontally to support the rafters.
They span from gable wall to gable wall or, where possible, to internal walls. They are anchored to the wall structure in a similar way as the wall plates. A safe method to anchor them is to join them firmly to the tie beam.
A joint is always a weak point in the beam. In purlins and ridge beams, therefore, jointing should be avoided wherever possible.
If a joint is unavoidable, its position should be carefully chosen. It should be placed in a position where the bending moment is small.
Rules for joints
a) Never use joints in purlins with only 2 bearing supports.
b) Purlins with 3 or more supports can be jointed at the bearing points.
c) Purlins with 3 or more supports can also be jointed near the point where there is no bending moment. This method takes advantage of the static system of a continuous spanned beam which has a reduced maximum bending moment compared to a single spanned beam.
Purlins and ridge beams; No joint possible
Design of joints
If the joint is at the place of the support, vertical halving joints (a) or scarf joints (b) can be used. Horizontal halving joints (c) must be avoided. Fish plate joints (d) are possible if the supporting wall is thick enough.
a) Vertical halving joint
b) Vertical scarf
c) Horizontal halving joint, to be avoided
d) Fish plate joint
If the joint is not at the place of the support, but where the bending moment is small, horizontal halving joints with bolts are used. The shear load must be transmitted by the bolt and not by the timber notch.
Design of joints; Correct halving joint
Correct and Incorrect halving joint; Rafters
The bearing area of beams must be sufficient to prevent too high a pressure across the fibres. The width (a) must be at least the same as the height of the beam (h).
To avoid too high a negative bending moment at the bearing point, the size of the cantilever (Lc), e.g. at verges, should not exceed 25% of the normal span (L).
Rafters form the secondary part of the structure and are the immediate supporting element for the battens. They are inclined in the direction of the roof slope.
To avoid too high a negative bending moment at the bearing point the size of the cantilever (at roof overhangs) should not exceed 35% of the normal span (L).
Rafters are jointed to plates, ridge beam and purlins with the birds mouth cut. To ensure a proper fixing, the width of the birds mouth cut should be half the width of the supporting beam.
The edges of purlins and wall plates should not be shaped because the rafters then would tend to slide down.
Birds mouth cut; Do not shape the corners of purlins
Rafters also have to be secured against uplift by wind. They are therefore fixed, to the under-structure either by proper nailing or with cleats.
Birds mouth cut; Do not shape the corners of purlins; Fixing with cleats
Nails for fixing must be long enough. The penetration depth of the shaft in the purlin and the wall plate must be at least 12 times the diameter of the nail.
At the ridge, rafters are jointed either by means of a plumb cut or a halving joint.
Fixing by nailing; Plumb cut; Halving joint
Lengthening joints should be avoided wherever possible. Where this is not possible, they should be near the point with the smallest bending moment. Fish plate joints are appropriate.
Battens are the last component of the structure. They directly support the tiles. They run horizontally and are fixed to the rafters by nails.
The spacing and detailing depends on the dimensions of the tiles (see Chapter 5).
The overhang over the last support at the verges should not exceed 30% of the rafters spacing.
The battens are laid flat on the rafters and not in upend position. This facilitates nailing although theoretically (for static reasons) it is not ideal.
Common practice: battens laid flat onto the rafters; In upend position not possible to nail
The nails have to be long enough. The penetration depth must be at least 12 times the diameter of the nail.
Joints are placed above the rafters, by means of a plumb cut.
In upend position not possible to nail
Trusses In truss structures great stress can occur, which cannot easily be determined with simple methods. The design of trusses, the sizing of their components and joints requires expertise, and their construction demands great care and professional workmanship.
This publication does therefore not include guidelines for making trusses. Specialist literature should be consulted.
4.4.1 Bamboo structures
Genera- Bamboo, as well as timber, is one of the oldest building materials and still used widely in many countries today. Its main area of distribution is the tropics. Some robust species are also found in subtropical and temperate latitudes.
Bamboo is a perennial grass. Well over 1000 species of some 50 genera are known. The largest number occurs in Southern Asia and on the islands between Japan and Java.
Bamboo differ from grasses in the long life span of their culms (hollow stalk) and in their branching and lignification (development of woody tissues). Like leaf-bearing trees, they shed their leaves annually and grow new branches, increasing their crown every year.
Bamboo is the fastest growing plant and has been reported to grow more than one metre in a single day. Bamboo culms can reach their full height (giant species grow up to 35 metres or more) within the first six months of growth, but it takes about 3 years to develop the strength required for construction; full maturity is generally reached after 5 or 6 years.
In many countries bamboo is a most important building material. Often it is not only available through commercial channels, but is also cultivated and used on a level of subsistence farming.
In roof construction bamboo can be used in the form of whole culms for building frame structures, trusses, beams, purlins, rafters; half culms as battens; and split bamboo strips for matting and woven panels, e.g. as ceilings.
With bamboo structures a completely even structure cannot easily be achieved. Therefore, in the case of FCR/MCR, only pantiles should be used and not Roman tiles or semi-sheets. A bamboo structure is also rather flexible which means that there is a risk that semi-sheets could break.
In many regions bamboo is abundantly available, cheap and is quickly replaced after harvesting, without the serious consequences known from the excessive use of timber (environmentally acceptable!). The annual yield by weight per unit area can reach 25 times the yield of forests in which building timber is grown.
Handling during felling, treatment, transportation, storage and construction work is possible with simple manual methods and traditional tools.
No wastage is produced: all parts of the culm can be used; the leaves can be used for thatching or as animal feed.
The pleasant smooth, round surface requires no surface treatment.
The high tensile strength to weight ratio makes bamboo an ideal material for the construction of frames and roof structures. With a proper design and workmanship, entire buildings can be made of bamboo.
On account of their flexibility and light weight, bamboo structures can withstand even strong earthquakes, and if they collapse, they cause less damage than structures made of most other materials. Reconstruction is possible within a short time and at a low cost.
Bamboo has a relatively low durability, especially in moist conditions, as it is easily attacked by biological agents, such as insects and fungi.
Bamboo catches fire easily.
The low compressive strength and impact resistance limit its application in construction. Wrong handling, bad workmanship and incorrect design of bamboo structures can lead to cracking and splitting which weaken the material and make it more vulnerable to attack by insects and fungi. Nails cause splitting.
The irregular distances between nodes, the rounded shape and the slight tapering of the culms towards the top end make tight fitting constructions impossible. Bamboo can therefore not replace timber in many applications.
Bamboo causes greater tool wear than timber.
Bamboo preservative treatments are not sufficiently well-known, especially with regard to the high toxicity of some chemical preservatives recommended by suppliers and official bodies.
Certain bamboo species have a natural resistance to biological attack, hence their cultivation and use should be encouraged.
Only mature and properly treated culms should be used. Bamboo should not be stored too long (if at all, then without contact with the ground), and carefully handled (avoiding cracks or damage of the hard outer surface). It should be used n carefully-designed structures (ensuring dry conditions, good ventilation of all components, accessibility for inspection, maintenance and replacement of damaged members).
Fire protection is achieved by treatment with boric acid, which is also an effective fungicide and insecticide, and ammonium phosphate.
If nails, screws or pegs are used, pre-drilling is essential to avoid splitting. Fastening of joints by means of lashing is more appropriate for bamboo constructions.
Bamboo should not be used where tight fitting components are required. Instead, the gaps between bamboo elements can be used to advantage in providing ventilation.
Recommendations for preservative treatments with chemicals should not be followed blindly. Different opinions of experts should be sought. Irrespective of the type of preservative used, care should be taken to protect the skin and eyes from coming into contact with it. The need for thorough safety precautions cannot be over-stressed.
Untreated bamboo deteriorates within 2 or 3 years, but with correct harvesting and preservative treatment, its life expectancy can increase about 4 times.
Priority should be given to non-chemical preservation methods such as correct harvesting and correct structural application.
a) Harvesting methods
Mature culms (5 to 6 years old) have greater resistance to deterioration than younger culms. Since fungal and insect attack increase with moisture content, bamboo should be harvested when the moisture content is lowest, that is, in the dry season in the tropics, and autumn or winter in cooler zones.
The freshly-cut culms, complete with branches and leaves, should be left standing for a few days (avoiding contact between the cut surface and the soil), allowing the leaves to transpire and reduce the starch content of the culm. This method, called clump curing, reduces attacks by borer beetles, but has no effect on termites or fungi.
b) Structural means
Similarly to timber structures (also see chapter 4.3.2), bamboo should be used correctly with respect to protection from moisture; access to circulating air; and avoiding contact with the ground.
c) Chemical treatment
Often harvesting methods and structural means are not enough to ensure a long life, making chemical treatment unavoidable. In this case attention should be paid to the same hints and warnings as for timber (see Chapter 4.3.2).
The chemical treatment of bamboo is more complicated than of timber because the outer skin of the culm is impenetrable. While split bamboo may be treated by soaking in a bath, whole bamboo culms require special methods:
Replacing the sap with a preservative solution, by allowing the solution to slowly flow from one end of the culm to the other, forcing the sap out. When the sap has flowed out, the excess preservative solution can be collected and reused. The process (called the Boucherie method) takes 5 days, but can be reduced to a few hours by pressure treatment.
Boucherie method, without or with pressure; Steeping
Immersing the lower portion of freshly cut culms (which with their leaves), in a preservative solution which is drawn up the capillary vessels by the transpiration of the leaves. This method (called steeping) only works with fairly short culms, as the liquid may not rise to the top of long culms.
Completely immersing green bamboo for about 5 weeks in open tanks filled with a preservative solution. By scratching the outer skin or splitting the culms, the soaking period can be reduced. Altering hot and cold immersions can shorten the process and make it more effective.
4.4.2 Pole timber structures
Pole timber is unprocessed round wood in the form of poles. It is one of the oldest and most valuable building materials. Timber poles are made from young trees (generally 5 - 7 years) with the barks peeled off, seasoned and treated as required (also see Chapter 4.3.2). The cost and wastage incurred by sawing is eliminated and 100% of the timbers strength is used. A timber pole is stronger than sawn timber of an equal cross-sectional area, because the fibres flow smoothly around natural defects and do not end as sloping grain at cut surfaces. Poles can bear great tensile stress around their perimeters resulting in high resistance during bending and compression.
Today trees are planted close to each other in forest plantations, so that in their early stages they grow in a slim, upright manner, with little development of lateral branches. These plantations are thinned out at regular intervals and as the cut trees are often of a suitable size, they can be used as pole timber.
Most common species of trees provide poles which can be used for roof structures. The more perishable species can be preserved to make them more durable. Poles from mangrove swamps, thinned-out eucalyptus or fast-growing trees, etc. are suitable.
Pole timber can be used for conventional roof structures consisting of purlins, rafters, ridge piece, wall plates etc.
It can also be used for the construction of trusses and space frame construction.
Pole timber structures are usually not absolutely even, similarly to bamboo structures. Therefore, in the case of FCR/MCR they are suitable for pantile covers; but not for Roman tiles and not at all for semi-sheet covers.
Depending on the type of structure, different jointing methods are used:
Simple structures can be jointed by cutting notches and nailing or bolting.
Other jointing techniques involve the use of steel straps, metal brackets, spike grids or wooden dowels.
Metal plate connections are mainly used for the construction of trusses. They are either simply wrapped around the joints and firmly nailed onto the timber, or are inserted into longitudinal saw cuts in the timber poles and connected to them by nails (flitch plate connection).
For frame structures a system based on special space frame
connectors can be used, comprising a cross-component of welded steel, and tail
end connectors with screws, washers and nuts.
A timber pole is stronger than sawn timber of an equal cross-sectional area.
A round pole possesses a very high proportion of the basic strength of its species because knots have less effect on the strength of naturally-round timbers, compared to sawn sections. The cost and wastage of sawing are eliminated. Above all, the design of any pole structure can be simple enough for unskilled persons to construct.
Pole construction makes use of less valuable timber such as young trees obtained from thinning out forest plantations. In this sense the technology is not harmful to the environment.
Pole timber structures are usually not absolutely even and are therefore suitable only for covers which are flexible, e.g. FCR/MCR pantiles, but not Roman tiles, semi-sheets and sheets.
Pole timber has a low-prestige image.
If pole timber is obtained by clearing young forests, this has a serious negative impact on the environment.
For testing the load bearing capacity of timber and sizing of basic elements of pole timber structures see Chapter 4.3.4 and Appendix 3.
4.4.3 Metal structures
Metal is usually a rather expensive material for building structures and is in most cases imported. Building with metal requires special tools and equipment. Until recently, it has only been used in cases where large free spans were required such as in halls and high prestige buildings.
Today, with the increasing scarcity of good quality timber, metal structures are becoming more and more a competitive alternative in roof construction.
Metals used in construction are divided into two main
Ferrous metals: irons and steels
Non-ferrous metals: aluminium (Al), cadmium (Cd), chromium (Cr), copper (Cu), lead (Pb), nickel (Ni), tin (Sn) and zinc (Zn).
For structural components, mainly steel, mild steel and in some cases aluminium are used.
Steels are all alloys of iron with a carbon content of between 0.05 and 1.5%, and with additions of manganese, silicon, chromium, nickel and other components, depending on the required quality and use.
Mild steels, with 0.15 to 0.25% carbon, are the most widely used and versatile of all metals. They are strong, ductile and suitable for rolling and welding, but not for casting.
Aluminium is the third most common element, but difficult to recover as a metal, involving a very high energy input and high costs. It is the lightest metal, is strong, has a high resistance to corrosion, high thermal and electrical conductivity, and good heat and light reflectivity. Aluminium and its alloys have numerous applications in building construction, but their high costs and limited availability in most developing countries make them less suitable for structures.
Structural steel, mild steel and aluminium are available in hollow forms (round, square or rectangular shaped pipes), rolled profiles and as sheets. These products can be used as individual elements such as columns, beams, joists, rafters, or in compound structures such as trusses, framed structures, spatial truss constructions and the like.
A combination of a steel structure and timber rafters and battens may be a possible alternative to save costs.
Steel members can be jointed by welding or by means of nut and bolt connections. Rivet connection is a technology which is no longer used in structures.
Corrosion is a typical problem of steel structures. An anti-rust paint is required, also on the welded joints. Parts that are exposed to moisture need periodical painting.
Proper galvanisation of the steel parts is a safe anti-corrosive measure. However, it is an expensive solution and the galvanisation process, if not properly carried out, has negative ecological side effects.
Galvanisation of welded structures is difficult in practice, because it should be done after welding and the galvanisation tanks are limited in size.
Aluminium does not require protection against corrosion. Welding aluminium is difficult, requires special equipment and skill. Therefore nut and bolt connections are usually used.
Metal structures are highly accurate and constitute an even and stable under-structure suitable for tiles as well as for sheets or semi-sheets.
Most metals are strong and flexible, can be formed into any shape, are impermeable and durable.
Prefabricated framed construction systems made of steel or aluminium are assembled extremely quickly. With strong connections, such systems can be very resistant to earthquake and hurricane destruction.
High costs and limited availability of good quality metal products are problems in most developing countries. As a result, inferior products are supplied, e.g. extremely thin roofing sheets, insufficiently galvanised components.
Metal structures lose strength at high temperatures and tend to collapse during a fire, sooner than timber, although they are non-combustible. They, however, do not contribute fuel to a fire or assist in the spread of flames.
Most metals corrode. Ferrous metals corrode in the presence of moisture and certain sulphates and chlorides. Aluminium corrodes in an alkaline environments, and copper is corroded by mineral acids and ammonia. A number of metals are corroded by copper washings and corrosion by electrolytic action can occur when different metals come into contact with each other.
Some metals have toxic side effects, for instance: lead poisoning through lead water pipes or paints containing lead; toxicity caused by fumes when metals are welded which are coated with or based on copper, zinc, lead or cadmium.
To prevent corrosion, use in humid conditions should be avoided, protective coating should be periodically renewed, and contact between aluminium and cement products (mortar or concrete) should be avoided.
Cost reduction can be achieved by limited use of metals and design modifications which permit the use of cheaper alternative materials. For instance, the combination of a steel main structure with timber rafters and battens may be a possible alternative.
Steel structure combined with timber rafters
4.4.4 Concrete structures
Concrete is a world-wide and abundantly-used material suitable for a large variety of applications. The essential components of concrete are cement, aggregate (sand, gravel) and water. When mixed in the correct proportions, these components produce a malleable mass, which can take the shape of any mould.
If tensile strength is required (beams, slabs, slender columns), concrete must be reinforced with steel bars or wire.
The main aspects that govern the quality of concrete are:
Careful selection of the type and proportion of the
Clean, hard and well-graded aggregate must be used.
Correct water to cement ratio.
Proper mixing and immediate use of the mix.
Sufficient compacting after casting.
Curing after demoulding for at least 14 days.
Concrete is either cast in situ or precast to be used in construction systems using assembly procedure. To increase the load-bearing capacity, the reinforcement can be prestressed before casting.
Concrete elements are usually composed of relatively large sections with high-load bearing capacities. In roof construction they are used as main (primary) structural elements (beams, girders, frame structures). Other materials are used for the finer structural parts (battens, etc.).
Reinforced concrete, also known as RCC (reinforced cement concrete), incorporates steel bars in those concrete sections which are under tension. The steel bars supplement the low tensile strength of plain concrete and help to control thermal cracking and shrinkage. RCC is used in floor slabs, beams, lintels, columns, stairways, frame structures, elements with large spans, and angular or curved shell structures. These are all cast in situ or precast. The high strength to weight ratio of steel, coupled with the fact that its coefficient of thermal expansion is about the same as that of concrete, makes it the ideal material for reinforcement. Where bars with ribs are available, they should be given preference, as they are far more effective than smooth bars. In this way up to 30% steel can be saved.
Prestressed concrete is reinforced concrete with the steel reinforcement held under tension during production. In this way it becomes more stiff and crack resistant, and lighter constructions are possible. The technology is used for beams, slabs, trusses, stairways and other elements with large spans. By pre-stressing, less steel is needed and the concrete is held under compression, enabling it to carry much higher loads. Prestressing is achieved either by pre-tensing, (when the steel is stressed before the concrete is cast) or by post-tensing (once the concrete has reached a specific strength, the reinforcing steel is passed through straight or curved ducts, which are filled with grout after the reinforcement has been tensed and anchored. This is essentially a factory operation requiring expensive, special equipment (jacks, anchorage, pre-stressing beds, etc.) and not suitable for inexpensive housing.
Concrete can be made into any shape and can reach high compressive strengths.
Reinforced concrete combines high compressive strength with high tensile strength, making it appropriate for any building design and all structural requirements. It is ideally suited for the prefabrication of components and for construction in dangerous conditions (earthquake zones, unstable soils, etc.).
Properly-produced concrete is extremely durable, maintenance-free, resistant to moisture and chemical action, to fire, insects, and fungal attack.
In many areas concrete has a high prestige value.
Cement, steel and moulds are usually expensive.
Quality control on building sites is difficult. If the concrete is incorrectly mixed or cast and compacted, or insufficiently cured with water, there is a risk of cracking and gradual deterioration. Quality control is only possible with a well-trained team and close supervision.
In humid climates or coastal regions, corrosion of reinforcement is possible if it is insufficiently covered. This can lead to expansion cracks.
Concrete is only fire resistance up to about 500·C. At this temperature steel reinforcement is no longer effective. After fire, RCC structures usually have to be demolished.
Demolishing concrete is difficult and the debris cannot be recycled, other than in the form of aggregate for new concrete.
The cement component in concrete can be reduced by a careful mix design, grading of aggregates, testing, quality control and by substitution with cheaper pozzolanas.
Savings in steel reinforcement can be made by good structural design and use of bars with ribs or prestressing with colddrawn low-carbon steel wire.