The systems are by no means universally applicable but in the right circumstances they are easy to install and very successful. Section 52 gives details of design.

This Section draws on experience of systems in the UK and USA that have been studied in detail by the author. The emphasis is on anecdotes, to illustrate use of diagnostics and practical difficulties of design and installation. Both internal and external sumps are covered.

In some houses, and especially in bungalows, small radon sumps can be the easiest and potentially most effective solution to a radon problem. Common features are a good depth of hard-core (often owing to the house having been built on a sloping site and where stone fill has been used) and with few cross walls beneath slab level.

In bungalows especially, installation has proven particularly easy where there is a full-height cupboard: the suction pipe can be located in a rear corner and the cupboard contents (often clothes) act as sound deadening material. The fan can be installed in the roof space, in a vertical position, and the outlet taken through the roof. This makes for a very neat design, with all components being hidden, and with all pipe-work within the inhabited volume being under suction.

Where deep hard-core layers exist (and often where they do not) large sumps are unnecessary. Several systems have been installed on a diy basis and for little more than the cost of a fan and a length of pipe. Where bungalows have shallow pitch roofs, obviously there is no need for scaffolding.

In houses, not only is there the complication of roofing work at higher level, but in many cases there is no easy and acceptable route for a 110 mm pipe to pass from ground floor to the roof space. Pipes passing through bedrooms may give rise to complaints about noise, and should preferably be located within wardrobes or airing cupboards, where sound deadening can occur. Use of proprietary ducting rather than 110 mm upvc pipe can ease installation problems but its cross-sectional area may be inadequate for situations where a high flow rate is predicted.

Pipe-work can suffer from external as well as internal condensation, and direct contact with clothing should be avoided except where the pipe is suitably insulated. Glass fibre insulation should not be used because any contamination of stored clothing could produce skin irritation.

Generally, sumps may be located near the centre of houses, or if diagnostics is undertaken, near to areas of high entry potential. The problem with locating sumps near to the edge of buildings is sometimes that there is excessive leakage into the cavity of an external wall, and much of the suction is dissipated. Just to complicate matters, this may be quite effective in limiting radon entry in some houses - but not in most.

In some cases depressurisation of foam filled cavities has acted to spread a ring of suction around the house, and has produced good results. However, there would probably be considerable heating cost penalties with these designs. Old houses are not immune to 'cavity leakage' where the construction is a thick wall with rubble fill. These old walls, common in rural parts of Devon and Cornwall especially, are often highly voided, sometimes inhabited by mice, and more than usually act to block the spread of depressurisation.

Floor constructions have been found to vary from 25 mm of cracked concrete (found still in many older houses in rural counties) to 400 mm of dense concrete. In the latter case, diagnostics enabled a solution not involving excavation of the floor.

The location of an indoor radon sump is often determined only partly by diagnostics: suitable locations of internal pipe-work or objections by the house owner to a certain room being disturbed can be overriding factors.

The difficulty of digging out modern well constructed concrete floors should not be underestimated: several builders have resorted to water cooled disc cutting equipment and complete room redecoration has had to be undertaken. In contrast, many old floors have been replaced with thicker concrete, and incorporating both damp proof membranes and radon sumps, as per the designs for new housing. Removal of the old floor may take less than an hour.

Complete new floors are of course invariably successful in reducing local radon levels, although problems of house depressurisation may occur because the suction extends effectively to slab edges, and to walls.

On more than one occasion it has been found that external as well as internal house walls had virtually no foundations. Use of nylon mesh material in place of hard-core may be recommended where excavation depth has to be kept to a minimum. In all cases the best performance is obtained with uniform size rounded hard-core, since this gives the highest void fraction, other factors being equal.

In houses where there is a large depth of hard-core, owing to a sloping site, it has proven practicable and inexpensive to insert a suction pipe into the hard-core from an outer wall, but beneath floor level. However, short circuiting via cavity leakage is almost inevitable, and although the systems often work well in radon terms, large fans need to operate at maximum power to produce sensible depressurisation even a metre away from the inlet, and noise from exhaust points has proved troublesome.

This is a particular concern in bungalows because the exhaust may be only a few feet above ground level. If pipe-work is taken to ridge height on a chalet bungalow, noise levels can be reduced, but the systems look 'decidedly odd', as it is unusual to see long lengths of pipe running up a gable end wall to above ridge height.

The aesthetic problems are exacerbated if, following design recommendations to limit condensation problems, fans are located at high level. Few householders may be prepared to tolerate such systems once the initial shock of discovering a high radon level dissipates. They may prove unpopular with Estate Agents too.

Additional aesthetic problems may be introduced by recent design recommendations to site the exhaust from radon systems at least 3 metres from any window or door or public path. The intention here is to ensure that none of the exhausted radon/air mixture can enter inhabited areas. Whether this advice is sensible or not depends on local circumstances: many houses have been cured of high radon levels with fan exhausts being within one or two metres. If the exhaust stream is expected to be exceptionally rich in radon (this can be predicted from diagnostics) more care needs to be taken.

The problems experienced with noise from exhaust points in high-flow sump systems are of course similar to those experienced with fans used to ventilate under timber floors, but here the added complication may be noise from the inlet end of the pipe. Systems have had to be modified to overcome these problems.

It has been emphasised in Section 52 that sump systems sometimes work poorly and occasionally do not work at all. Usually these failures may be traced to very low permeability ground, cross walls, failure to undertake entry potential diagnostics, or (at the other end of the scale) to an infinite source. It has proved interesting to resolve some of these problems: the usual solution of "put on a larger fan" may not produce adequate results and noise levels may increase.

In non-domestic buildings multiple fans have been used, again sometimes not to good effect.

It should never be forgotten that pressurisation may offer an easier solution to difficult houses than can multiple sumps, and effort devoted to digging out floors might be better spent in sealing up the house to an extent that would permit successful operation of a pressurising system. Consultancy advice should be obtained in these cases.

The limiting factor may be leakage where joists or hangers are set into walls at first floor level. In the most difficult cases, each house has to be treated on its merits, and a full range of diagnostics may be necessary. Thankfully, such houses have proved to be rare in the UK. The usefulness of diagnostics on simpler houses also is that it can indicate design solutions that entail less disruption and greater effectiveness than more obvious remediation.

Most radon houses respond well to standard treatments, but many might have been mitigated to a better extent were adequate diagnostics to have been utilised. However, here the issue is one of cost and marginal cost effectiveness: if an 80% cure can be achieved simply and easily, it has to be questioned whether it is worth the cost and necessarily uncertain outcome of extensive diagnostics even if some rooms remain over the 200 Bq/m³ level. (It is usually easy to achieve less than 200 Bq/m³ in at least one room!)

It is a problem defining the radon level in properties mitigated by sump systems: often the variation room to room may be greater (at least in percentage terms) than before remediation, and with some rooms above and some below the 'action level', but depending upon the period of measurement and obviously upon window opening habits. At the present time, and especially in view of measurement uncertainties (some of which can never be removed because they derive from building imponderables) it would be more honest to ascribe to radon results a degree of uncertainty, rather than to quote blandly in terms of two or three significant figures.

External sumps.

The performance of external sumps in UK housing has proven very variable. Some systems work well and for minimal cost and disruption whilst others do not work at all. Communication testing is the first step of diagnostics, but can prove frustrating. Systems that 'ought' to work do not, because of the presence of high entry potentials that are unaffected by the system. Also, it is not unknown for powerful fans slightly to depressurise a house via wall leakage.

Detailed design of external sumps is important so that the pressure field can extend as far as diagnostics would suggest should be possible, and to avoid cavity short-circuiting - which has proved troublesome. Supervision of the works by a consultant may be recommended in view of the highly variable performance and sensitivity to design parameters and small constructional differences. Similar comments apply of course to side-entry sumps on sloping sites.

The importance of one design parameter may have been underestimated. A large area of concrete or paving around a house may exacerbate the indoor radon levels, but can help facilitate good performance of an external sump. Much then depends on wall design and construction.

These problems are in sharp contrast to early successes reported from Sweden. There, a group of houses built on radium rich but highly permeable ground were apparently cured of their radon problems by using a single large fan and a single deep sump dug tens of metres from the houses.

The key factor would be the permeability of the ground (which would have exacerbated the radon problem for a given source activity concentration) and the good communication across many houses. The cure may be seen in terms of ventilation of an area of ground, but without the complication of a concentrated source of the type that can prove troublesome if located beneath a house or other building in a mining district. Similar successes in the UK seem unlikely in view of ground conditions in many affected areas.

next page

top of radon section