Ground penetrating radar was first used to investigate concrete structures more than 25 years ago, initially to look for voids under concrete footpaths. The basic principles have remained largely unchanged, even though there have been massive advances in associated technology such as control systems and software, and in the understanding of how electromagnetic pulses behave inside concrete elements.
These days, radar pulses are very short - between 1 and 10 billionths of a second - and have frequencies between 500MHz and 1.5GHz. Energy losses within the concrete limit the total travel path to between 10 and 20 wavelengths, and the dimensions of distinct layers less than half a wavelength thick cannot be detected.
Thus, pulses of 1GHz with a wavelength in concrete of around 100mm can penetrate to a depth of about 500mm and measure layer thicknesses of 50mm and above. However, the location, but not the dimensions, of a 10mm thick waterproof membrane could be recorded.
This is in air-dry, well-compacted hardened Portland cement concretes. Moisture in the concrete has a significant effect on radar performance, and concrete generally has to be at least three months old before radar is an effective analytical tool.
So says the recently-published Concrete Society Technical Report No 48 Guidance on radar testing of concrete structures . It sums up the use of ground penetrating radar to investigate concrete structures thus: 'At the current stage of its development [GPR] is most readily used as a rapid 'sweep' or reconnaissance technique, from which an overall assessment of quality and condition may be obtained.'
This is because 'the combined and localised effects of changing moisture content, density and aggregate concentration throughout a concrete unit, to mention only one set of factors, will ensure that absolute measurement of all engineering properties is not possible, and that correlation with physical methods of measurement will always be necessary as an accuracy check'.
Professor John Bungey of Liverpool University, the Concrete Society's radar testing working party chairman, adds: 'What radar does best is the comparative identification of element thickness changes and the location of metallic objects such as pipes and reinforcing bars at depths which may exceed the normal range of covermeters.
'Where it is least effective is attempting to locate small or complex non-metallic features which lie close to the surface of the concrete.'
According to the report, air voids with less than 50mm plan dimension are unlikely to be detected, and reinforcement bars at 100mm centres or less are likely to mask deeper features. It is limitations such as these that cause Bungey to warn against 'over-enthusiasm' among those promoting radar techniques.
He adds: 'Clients must be clear in their own minds exactly what they wish to achieve - including the level of detail desired - and communicate this effectively to the radar survey contractor. This will permit a realistic judgment as to the suitability of the technique and the selection of the most appropriate equipment.'
One key point made by the report is that an X-ray type image is not yet possible. 'A radar survey produces signal patterns which require interpretation by an experienced person, and then, perhaps, review by an experienced engineer, of the effect of the patterns found on the characteristics of the tested unit.'