The subject of radon diagnostics is complicated by five factors:

the range of equipment available,

the claims and counter claims for this equipment,

general weather conditions, which can lead to little radon being found in a known high-radon house over periods as long as a few days,

wind direction, which can much influence the radon distribution within a house or other multi-cell building

the simple (but unwelcome) fact that what may prove useful and conclusive in one house may be a waste of time in the next.

Often tests are designed to suit the building and what is required to be known about it. This is particularly the case with ad-hoc tracer gas studies.

A good starting point is that tests should always be undertaken with a clear idea of what will be the usefulness of the result, assuming that a representative result can be obtained on the day in question. What comprises a representative result may of course not be known!

Diagnostics falls conveniently into several distinct sections:

Distribution of internal radon concentration

Air tightness of the building

Pressure differentials across floor slabs and suspended floors

Depressurisation of rooms by fires, fans, and radon systems

Air flow into buildings via uncharted pathways

Assessment of underfloor conditions: radon concentrations

Assessment of underfloor conditions: entry potentials

Vacuum suction testing.

Distribution of internal radon concentration.

It is an unfortunate fact that both short term and long term radon concentrations in UK high-level radon houses are far from constant or reproducible. There is less experience in moderately affected houses, but limited data supports a high degree of variability here also.

There are several factors that conspire to produce this situation:

UK houses often have a mix of floor constructions and occupancy patterns of each room

UK houses are almost invariably naturally ventilated

weather conditions can have a marked effect on short term (days) and instantaneous radon levels

many buildings in affected areas are old, and have been extended and altered over the years. They are now effectively two or three different buildings, connected together and occupied by a single household.

Some houses are remarkably constant in their radon levels, others are remarkably variable. To some extent this can be explained by local source terms that are highly weather dependent. In other buildings, wind speed (for example) appears to have little effect. Some houses known to the author can almost be guaranteed to be free of radon in a high wind. Others can be guaranteed to exhibit high indoor levels. Only when the behaviour of each building is well understood can such effects be predicted.

Long term variations in

indoor radon concentrations in given rooms may exceed a factor of 4, without action by occupants or radon specialists. Long term average levels in different rooms in the same house can vary by up to a factor of 40 (see Section 60), and short term levels by almost any factor.

The most constant levels room to room are found in houses having connected timber suspended floors. In these cases the underfloor space forms an effective mixing zone, analogous to the full basement in a typical house in the northern USA. Variations room to room over long periods of time (weeks, months) are typically less than 2. Factors in excess of 3 appear to be unusual.

The most extreme variations are seen typically in houses that have been extended over the years, and where there are different floor types. For example, an old house with part timber floors over an old badly ventilated cellar or crawl space may have added to it an extension with a good quality concrete floor. There is no guarantee of where the highest radon levels will be found but the suspended floor areas are sometimes the worst affected and the most difficult to remedy.

These factors need to be recognised in deciding how to measure radon concentrations and over what time period. If equipment that detects radon daughters is used there are added complications. In free room air, the equilibrium factor (see Section 7) is usually between 0.3 and 0.6. In small spaces however, such as in under-stairs cupboards and beneath baths, the surface/volume ratio is high and radon daughter readings may be much lower than expected for a given radon gas level probably owing to enhanced plate-out.

Recommendations:

In houses with timber floors throughout a single measurement point may adequately characterise the radon level in ground floor living areas if averaged over several weeks and if no abnormal use is made of the windows in the chosen room. Floor coverings such as carpet or vinyl may have only limited effect, except where the vinyl is very well fitting. A departure from the true whole-floor average of more than a factor of 2 or 3 is unlikely. Nevertheless, day to day variations may be extreme.

In houses with solid floors throughout a couple of measurements may be advised in ground floor rooms, especially if rooms have either a different age of floor or are built into a hillside. In extreme cases variations room to room can be a factor of 6 or 7, but 2 or 3 is more usual. The information from these measurements taken over weeks can be used to 'correct' longer term screening results taken only in one of the rooms, and for the purpose of determining average annual levels.

In houses having mixed floor types and especially if of different ages, multiple measurements may be advised, preferably under the control of a consultant if the house is thought to be badly affected. Room to room variations may be extreme, as may day to day variations.

All measurements to determine room to room variations should preferably be taken over a week or more, and it is likely that electret detectors will prove the most popular. Track-etch detectors are insufficiently sensitive for short period exposures, and charcoal canisters cannot average for more than a few days. Nevertheless they can prove useful.

Measurements using 'active' equipment that can produce a read-out of radon level in each room within 10 to 20 minutes must be used with some care. It is sometimes the case that short term measurements show a similar room to room variation as longer term measurements but this cannot be guaranteed, even when the building has been closed up prior to the test period. The reasons include that a closed up condition is atypical of usage, and that wind direction and strength in the hours before a test can markedly influence the respective levels in each room in some houses.

Measurements to determine the long term average radon level in a house should, as is accepted practice, be taken over months, and if possible between autumn and spring. Summertime results can be deceptive, especially in some old houses. Track etch or electret detectors are the most suitable passive devices, with electrets being perhaps less prone to end (background) errors and calibration errors. However, different types can be affected by gamma emissions. Instructions should be followed in all cases.

Air tightness of the building.

This parameter is important for advising on ventilation of a house, and for predicting the likely effectiveness of positive pressurisation systems. When undertaken as a part of research studies a 'blower door' may be utilised. These consist essentially of a large fan set into a framework that can be adapted to fit into a range of door openings. Fan speed can be controlled and the system can operate to pressurise or depressurise a building.

Under suitable weather conditions parameters of a standard air flow equation can be obtained, and the house or other building classified as 'leaky' or 'airtight' as appropriate. In general, UK houses are less airtight than many in Scandinavian countries - a consequence of building construction standards and the greater benefits of energy efficient design in colder climates.

For radon remediation purposes, use of a blower door may be considered unnecessary. Simpler systems based on a commercially available vacuum cleaner or fan can be utilised to introduce air through the letter box or other suitable opening, and the resulting pressure difference across the house structure can be measured using proprietary equipment.

Measurement of flow rate can be by a number of techniques including orifice plates and pitot tubes, but with care to observe the usual precautions when using air flow transducers. By these means, it can rapidly be determined whether a house is likely to be cured wholly or in part by a positive pressurisation system. The effects of sealing selected large openings can also be determined.

In effect, it can be arranged that the test fan introduces into the house a flow rate of air at least equal to what would be introduced by a pressurisation system, but having regard to the pressure/flow characteristics of the proprietary system.

In most houses, even the existing kitchen or bathroom extract fan may be utilised to provide sufficient air flow for test purposes, but the flow will be out of and not into the house. The author's record here is a depressurisation of nearly 15 Pa by a small kitchen fan, but 30 Pa has been known.

One advantage of depressurising the building is that large air leakages into the structure can be found with 'smoke sticks' or 'smoke guns'. Favourite pathways are under sinks, under baths (in bungalows), and where other service pipes and cables enter from the outside. In many houses however, large flows can be detected using no more than a wet finger.

Pressure differentials across floor slabs and suspended floors.

It is important to recognise that these pressure differences are near the limits of measurement using ordinary test equipment. Units are pascal (Pa), and one inch of water column is equivalent to about 250 Pa. Pressures across floors are typically 0.3 to 3 Pa. Sump pressures are in the range 100 to 250 Pa.

Almost always the most suitable test equipment is an electronic micromanometer. These can be obtained from several sources in the UK. Care must be taken in using these instruments because they can easily be damaged by over-pressure. For averaging purposes, either electronic integrators or a chart recorder may be used, provided that the manometer has a suitable output.

Most of the difficulties arise because wind pressures can be greater than the pressure difference under test. Thus, many days (especially in Cornwall) are entirely unsuitable for testing. Sometimes, it is desired to determine whether suction spreads across a floor slab, and whether it remains sufficient to balance the stack effect - usually about 2 to 3 Pa in a two-storey building.

Depressurisation of rooms by fires, fans, and radon systems.

 The equipment used for these tests is similar to that used for across-slab measurements, but with more care being necessary to consider where the reference pressure should be taken. For low pressure differences, it may be important to ensure that the correct difference is being measured: a room to room measurement may be different from a room to underfloor measurement, especially in windy conditions. Where possible a reference pressure from a sheltered location (ie, not directly outdoors) should be selected.

Open fires have been known to induce depressurisation as high as 15 to 20 Pa, and kitchen fans can manage 20 to 40 Pa in extreme conditions. In contrast, radon systems may depressurise the building only by a small fraction of a pascal, but this can be sufficient substantially to alter average flows around the house, and from some minor radon entry routes.

The reason why these small pressures can be significant is that they occur for 24 hours per day, rather than the few minutes or hours per day that is more typical for extract fans and fires. Their measurement is often difficult, and on/off operation of fans is essential. Output should be to an integrator or chart recorder, and the tests need to be undertaken with all windows closed and with no people moving around the house. Calm weather conditions are essential.

Air flows into buildings via uncharted pathways.

These tests can range from difficult to impossible, and much may depend on how much dismantling of the house can be tolerated. Nevertheless, often telltale signs of air entry from curious locations can be observed without test equipment. Classic cases involve stains around the edges of bedroom carpets: air can flow up cavity walls to enter a house via openings around joist hangers or joists. Where a thick white carpet is fitted close to the skirtings, it is not unusual to observe marked staining, indicative of the carpet having acted as an air filter.

Whether these air entry routes are significant in radon terms will depend upon the average radon concentration in the cavity, and average flow rates, and these will depend on a host of constructional details, including whether the house is rendered. (Rendering can reduce the air flow into and out of a wall, and may inhibit clearance of radon via alternating wind pressures.)

Tests of radon concentration in cavities can produce remarkable results, with quite different readings being obtained at different points, indicative of considerable air movement within the cavity or localised sources. Wind conditions can affect these readings to an alarming degree: sometimes cavity levels can be higher than in adjoining rooms (indicative of the cavity as a possible source) whilst on different days tests in exactly the same locations can produce contrary results. The usual glib explanation in terms of 'wind effects' disguises that little is known about how air moves from the ground into cavities and from there into and out from buildings: it would be a brave scientist who would predict on-site behaviour given only the design drawings of the house.

Assessment of underfloor conditions: radon concentrations.

Many types of proprietary equipment can be used to measure radon levels in underfloor spaces and within cracks in floors. These can be classified either by technology or according to whether they measure air in bulk or whether they can sample a small volume. This is important, because often the act of taking the sample can markedly influence the result obtained.

Manufacturers catalogues should be consulted for up-to-date details of equipment. Amongst the most popular types are a range of devices that utilise scintillation flasks and photomultiplier tubes to enable measurement of air samples drawn from room air, underfloor air, or air within cracks. The flasks may be used in 'flow-through' or 'grab sample' modes, the latter being almost universally applicable for on-site investigations.

When measuring radon levels within a room or in an underfloor space there is little problem with accuracy as taking the sample does not much influence the source. However, when sampling from a crack or behind a skirting board, the result may be much influenced by the ratio of the effective volume of the air space being sampled to the volume of the sample. Without dismantling the building it is often not possible to determine this effective volume. Thus readings must be interpreted with care, and can form only a rough guide to radon entry routes. The results are meaningless in absolute terms, because even sticking tape over a length of crack can increase a result by a factor of 3 or 4: the explanation is simply that more of the sample is drawn from radon-rich air behind the skirting board, and less from the air in the room.

These effects can introduce considerable confusion if it is attempted to determine by grab sampling which end of a room is 'worst affected' by entry routes. However, it is easy enough to determine whether the sample space is small (and the reading variable) or essentially infinite.

It has to be appreciated that soil gas concentrations of 10,000 Bq/m³ are not unusual. Concentrations of 50,000 Bq/m³ behind a skirting board indicate a source for further contemplation, but only readings in excess of 100,000 Bq/m³ indicate extremely active ground. The authors record reading for Cornwall is 1,200,000 Bq/m³, and was obtained beneath a lounge floor in Redruth. Curiously (see Section 60) the lounge was not excessively high in radon despite the poor state of the old concrete floor: far worse had been experienced elsewhere and over less active ground.

It is important to recognise that the average indoor radon levels in a building do not scale well with underground radon concentrations: so much depends on permeability and use of the building. Some of the worst affected radon houses in the UK have underfloor radon concentrations that seem never to exceed 20,000 Bq/m³, but when combined with an 'infinite source', cracked floors, double glazing and draught-proofing, it is not difficult to accept annual average indoor levels in ground floor rooms in excess of 4000 Bq/m³.

Radon levels can also be determined by equipment that draws large quantities of air through filters or into chambers. These instruments are unsuitable for many diagnostic purposes, yet are sold as 'universal' measurement devices. Other instruments are based on ion chambers, but in the authors experience can be unsatisfactory.

Assessment of underfloor conditions: entry potentials.

In an ideal world, radon systems would be designed having regard to a map of radon entry potential of each building. Radon entry potential is the product of the flow rate and radon concentration that can be maintained at any point using a pressure difference typical of the region in which the point is located, and under representative weather conditions. In the real world, measuring entry potential can be difficult under on-site conditions. Also, the techniques are too involved for commercial companies to use since they can involve considerable time on site and investigation of each room in a house. Nevertheless they are useful for research purposes and for consultancy investigations of difficult buildings.

Radon enters buildings because of small pressure differences across floors and other boundaries. In radon potential testing, a small pressure difference is applied using an air pump (or a throttled vacuum cleaner, or a kitchen fan or a blower door) and the flow rate measured. The maintained radon concentration is determined by sampling the air, and the product gives a good indication of the extent to which that part of the underfloor area would be capable of sustaining a radon entry into the room under normal conditions. Of course, the entry potential from the ground has to be matched by entry routes into houses, but often the main flow resistance is the ground itself. Thus the technique gives reasonable answers where applied to typical 'slab on grade' houses built on low permeability soil and without a hard-core layer.

The usefulness of the technique is indicated in the following example:

Where underfloor conditions are those of an infinite source (an old mineshaft) large flow rates are needed to produce measurable pressure differences across the floor. Maintenance of the radon level under these conditions confirms the diagnosis. If the radon reading drops markedly, this indicates not a high entry potential but a substantial short circuit from the test point to outdoor air. Simple testing only of the radon concentration in the absence of flow might produce a similar result in these very dissimilar cases.

Few houses in the UK have been investigated in the detail necessary to explore the full potential of the technique, or more correctly, few householders have been persuaded to tolerate the author's extended experiments.

Also, different results may be obtained on different days: during windy conditions, the source term, as represented by the radon concentration in the subjacent ground, may be significantly reduced.

This can be a severe problem when testing houses built into hillsides, because wind can literally blow the radon 'clear away', leaving no evidence of high levels within all the usual cracks and gaps around the floor. Radon entry potential testing can therefore give a misleading result, even to the extent of a researcher wondering if he might be investigating the wrong building, so marked can be the discrepancy between entry potential results and notified long term radon levels.

The remaining difficulty is that the the relatively constant depressurisation that can be imposed by blower doors may not represent how the house behaves in use, especially if open fires are used. Room to room potentials may not match well with notified average radon levels. To some extent mitigation systems can be designed to suit the use of the house as well as its innate characteristics.

Vacuum suction testing.

An established technique is to test the applicability of a radon sump system by mimicing the performance of the system using a vacuum cleaner. The aim is to test the extent of pressure field distribution from the chosen sump location. The test has limited validity in respect of determining whether a system will adequately reduce radon levels because of the number of other factors involved, but is useful in characterising sub-floor communication, especially where there are cross-walls, and in selecting the most suitable type and size of fan given the flow/pressure readings obtained from the suction test.

Vacuum suction testing is often termed communication testing. Poor communication is indicated by regions of maintained positive pressure (or zero pressure) when the simulated sump pressure is set between 100 and 200 Pa, typical for an in-line centrifugal fan. Obviously the flow rate can be measured at the same time, and the technique then becomes somewhat akin to 'entry potential' testing.

next page

top of radon section