Repairs Where Leakage is Inwards Into the Structure

So far in this section consideration has been given to the repair of liquid- retaining structures where the leakage is outwards from the structure. Leakage or seepage into the structure can take place with both liquid- retaining and liquid-excluding structures. In the former type the in filtration would only occur when the pressure outside exceeded the pressure inside, in other words when the structure is partially or completely empty. When infiltration does occur, then this can be very difficult to cure completely, because it's usually impossible to carry out the repair from the water (outside) face.

9.5.1 Repairing Leaks in Basements

Consideration of this problem includes basement structures in which special electronic equipment is fixed and no moisture penetration can be tolerated. In cases of this category, it's usual to tank the structure, i.e. provide a complete waterproof membrane under the whole of the floor and carry it up the external walls to above ground level. Even so, the problem of ensuring that such a large area of membrane is completely watertight and remains so during the construction of the structural floor and walls and for the lifetime of the building is a formidable one, and is not always successful.

It is obvious that at the design stage, detailed consideration should be given to the standard of watertightness required, bearing in mind that good quality, well compacted concrete is watertight but not vapor-tight. The standard required for the sump of a sewage pumping station would be different to that of an underground car park, which in turn would differ from a basement used for storage of materials liable to be damaged by high relative humidity.

When the specified standard of watertightness has not been achieved and repairs are required, the problem is to decide how these repairs should be carried out, taking into account the circumstances of each case as far as they are known. The words ‘as far as they are known’ are important because in most cases nothing is known for certain about the actual condition (impermeability) of the concrete which is behind the inner face of the wall. All that's known is that water penetration is occurring; this may be along joint lines or it may be in random areas of the floor and walls.

In the case of penetration on joint lines, this may be due to a fractional opening of a ‘monolithic’ joint which does not have a water bar and /or sealing groove on the outside. If there is a water bar then the seepage indicates either some displacement of the water bar or honeycombed (under-compacted) concrete around the bar. In the random areas, this is due to under compacted concrete.

What is emphasized here is that the extent of the under compaction (or degree of porosity) is not known, unless cores are taken or an ultrasonic pulse velocity survey is carried out. It is the author’s experience that it's very unusual for either of these two methods of investigation to be adopted. It is general practice to stop the leaks by sealing the inside face of the floor or wall. There are a considerable number of proprietary materials on the market in this country, the Continent, USA, etc., which, when properly used, will seal off the inflow of water. Most of these materials are fairly new inasmuch as they have been developed over the last 15 - 20 years. Their durability in terms of the normal lifetime of a reinforced concrete structure is therefore not known for certain. This statement is not intended to suggest in any way that these or other new materials should not be used. If this attitude were adopted, there would be no progress at all in the development of new materials and techniques.

Questions are sometimes asked whether the water penetration and consequent repairs will result in increased maintenance costs. It is the author’s opinion that it's reasonable to assume that some additional maintenance expenditure will be required compared with the same structure if no water penetration had occurred at all. However, it would be quite unrealistic to assume that a large reinforced concrete basement can be constructed in a subsoil which contains a water table, without any water penetration at all. It can be done in theory, but not in practice unless the basement is tanked. From this a fundamental question arises, namely, what effect will water penetration have on the long term durability of the structure from the point of view of structural stability and maintenance. The author has already given his opinion on maintenance costs, but he feels that detailed consideration of the effect of water penetration would be useful.

The reinforced concrete walls of basements are either (a) load-bearing or (b) panels spanning between reinforced concrete columns. The floor slab is usually uniformly supported on the ground with additional reinforcement to take any uplift due to water pressure, or is a suspended slab, also with additional reinforcement. The first matter to be considered is whether the ground water will attack the concrete itself. Some ground waters are mildly aggressive to Portland cement concrete, but not sufficiently so as to cause any significant attack on an adequate thickness of high quality, dense, impermeable concrete. When sulphate- resisting Portland cement is used, it's always emphasized that a good quality well-compacted concrete is also required, so that the sulphate- resisting properties of the cement will operate to the best advantage.

It has been stressed several times before in this guide that steel embedded in Portland cement concrete is protected from corrosion by the intense alkalinity of the cement paste. In other words the steel is passivated. Unless this passivation is broken down either by a reduction in the alkaline environment provided by the cement paste or by other factors such as the presence of chloride ions, corrosion of the steel will not occur.

If the leak is sealed off on the inside face of the wall or floor, this does not prevent water entering the concrete from the other side, but it does stop any flow-through. This prevention of flow and the establishment of static conditions is more important than may appear at first sight. Concrete subjected to continuous water pressure will in the course of time become saturated. The rate at which the water will penetrate the concrete will depend on the permeability of the concrete. With high quality, dense, well compacted concrete the permeability rate will be very low. This very slow passage of water into the concrete will not, as far as is known at present, result in deterioration of either the concrete or the steel reinforcement, unless the water contains aggressive chemicals in solution. Also, for corrosion to occur, a supply of oxygen is required at the surface of the steel. The amount of oxygen present below ground level is much reduced compared with the open air.

Clearly, the actual amount of water which penetrates into the concrete is an important factor. Unfortunately in the type of structure considered here, this factor is not known. If for example, the concrete in the outer part of the wall or lower part of the floor slab were very porous, there may be so much water penetration as to reduce the alkalinity of the cement paste below the level required for effective passivity and then corrosion of the outer (or lower) bars may occur. However, unless the areas of seepage were considerable, it's unlikely that the strength of the wall or floor as a whole would be affected to any significant degree.

From this brief discussion it's obvious that each case of water penetration must be considered on its merits. Surface sealing of walls and floors on the inside face, remote from the point of entry of the water, has been successful in providing a reasonably dry basement. It can therefore be considered as an established and accepted method of repair. An alternative or additional method is pressure grouting, which has been briefly described earlier in this section. It can't be relied upon to completely seal off areas of infiltration, but if carried out by experienced contractors, it will greatly reduce the penetration of water. It has the advantage that the grout will penetrate into the sections of concrete which are honeycombed, thus providing direct protection to the reinforcement in that part of the wall or floor.

It is the author’s opinion that a specification for a new structure of the type considered here should contain clear directions on how any necessary repairs to prevent ingress of water should be carried out. Such a specification may require both pressure grouting and surface sealing when the wet area or amount of moisture penetration exceeds stated figures as this would help to ensure long term durability.

9.5.2 Repairing Leaks in Roof Slabs of Water-retaining Structures

In some structures, minor leaks in the roof may not be viewed with much concern apart from the danger of corrosion of the reinforcement. However, in the case of drinking water reservoirs any leak is a potential source of contamination. Joints and cracks in the concrete slab are likely to be the principal cause of leakage. Porous concrete may contribute, but is seldom the main cause of the trouble.

It is now normal good practice to provide a waterproof membrane over the whole of the roof slab, but this was not the case some 30 years or so ago. The absence of the membrane is often accompanied with inadequate falls to the slab. This results in ‘ponding’ and may lead to the gradual saturation of the concrete. In the course of time the alkaline environment around the steel may be reduced to such a level that corrosion occurs.

Where there are no definite leaks, but rust stains on the soffit of the slab, it's very difficult to decide whether the corrosion is due to the porosity of the concrete cover, i.e. the soffit concrete, or to penetration of water from above.

The earth cover, if any, must be removed, and the surface of the concrete thoroughly cleaned, with special reference to joints and cracks. In repairing joints and cracks, the major decision has to be taken on the method of repair, whether to use a rigid or flexible material. With uncovered, exposed slabs, the temperature range may be as much as 70°C, while with a cover of 300— 400mm of earth, the range may not exceed 15°C. With a large temperature range, the joints and /or cracks are likely to open and close seasonally and therefore flexible sealants should be used. With a much smaller temperature range, many of the joints and cracks can be safely repaired with a rigid material, so as to ‘lock’ the joint or crack.

Many older structures have no purpose-made movement or partial movement joints in the roof slab. This often results in some of the construction joints opening and forming what in reality is a number of stress relief joints. Cracks are sometimes formed by the same cause. Where there is a definite leak through a joint, a practical way to effect the repair is to remove the whole of the existing sealant and replace it with new. If there is any inert filler in the joint this should also be renewed. The new sealant can be any of the materials described in Section 1, bearing in mind the characteristics of the various types. If a preformed gasket is selected, the width of the gasket must be wider than the groove into which it will be fixed, and the use of the correct ‘oversize’ is essential if a watertight joint is to be achieved. Figure 9.15 shows the use of Hypalon sheet and an insitu sealant. If a crack is sufficiently straight it should be possible to seal it with the method shown. Unfortunately cracks are seldom straight, and then the only practical method is to cut it out with the special tool.

In structures built in the 1920s and 1930s, it's sometimes found that the roof slabs are reinforced with XPM instead of round bar reinforcement as is the practice today. The author has found that quite severe corrosion of XPM can occur without spalling or cracking of the concrete. Rust stains are visible and sometimes the outline of the XPM can be clearly seen. When this happens the concrete should be removed to allow the XPM to be examined and an assessment made of its value as tensile reinforcement. If the XPM is seriously corroded, it's advisable to provide new reinforcement on the soffit of the slab, anchored into the beams and properly gunited in. At the same time a decision should be taken as to the cause of the corrosion, i.e. penetration of water from above or below. In practice it may not be possible to arrive at a clear cut answer to this, and in this event, it would be prudent to provide protection on the top surface and on the soffit of the slab.

If the suspected seepage is widespread, the provision of a new fully bonded waterproof insitu membrane of polyurethane or preformed sheeting would be justified. An alternative is the use of unbonded sheeting. The advantages of this latter method are that the sheets, being unbonded, are not subjected to strain due to movement of the roof slab; also, they can be laid in almost any weather.

FIG. 9.15. Method of sealing contraction joint in roof slab.

However, if the unbonded sheets become damaged and water can pass through the holes or tears, then it will gradually flow over the roof slab until it finds a weak spot and then will penetrate the slab. With fully bonded sheets this will not occur as long as the adhesive remains intact.

Roofs of reservoirs are sometimes very exposed and it's essential that the unbonded membrane be held down against the suction which develops during periods of storm and strong wind. This is usually done by carefully spreading rounded (not angular) shingle on the surface to a depth of about 50 mm.

The sheeting is in fact fixed to the perimeter of the roof and is carried up and fixed against parapet walls, pipes and other members which pass through the roof slab. The sheets are lapped and solvent (cold) welded and then finally sealed with a special material. Detailed information on the use of insitu coatings and preformed bonded sheeting for lining reservoirs and similar structures has been given earlier in this section.

Other materials which can be successfully used to hold-down and protect unbonded sheeting, include no-fines concrete 75 mm thick, and 50 mm thick precast concrete slabs. These two materials can also be used to provide a protective cover to insitu coatings and bonded preformed sheeting, capable of taking foot and light traffic.

There will be cases where a membrane has been provided, but in spite of this, the roof leaks. The location of the defect(s) in the membrane can be very difficult if not impossible because it's extremely unlikely that points of visible leakage on the soffit of the slab will coincide with the defects in the membrane. The decision on the best method of repair will depend largely on the extent of the leakage. If the leakage is extensive probably the most satisfactory repair will include the removal of the existing membrane and its replacement by new material. At the same time the opportunity should be taken to seal the defects in the concrete slab. If the leakage is relatively small it can be repaired on the underside of the slab by one of the methods described in the previous section.