Reusing foundations demands exceptionally detailed site investigation. Justin Phillips reports on work at Battersea Power Station.
London's Battersea Power Station was decommissioned in 1982. Plans to turn it into an indoor theme park foundered but the site was given a new lease of life in 2001 when the present owner, Parkview International, gained detailed planning consent for a mixed development forming one of the UK's largest brownfield urban regeneration projects.
The Grade II listed power station building will remain intact.
Internally, the space will be transformed to become a premium retail and leisure destination complemented by activities and experiences based on the theme of design excellence.
Two new hotels will be housed on the surrounding site - a 400room five-star urban resort and a 750-room conference facility, one of the largest in Europe.
Before any detailed design work began, Buro Happold Ground Engineering carried out an extensive ground investigation, including an assessment of the potential for reusing foundations within the power station.
The site's rich history includes a legend that the Duke of Wellington fought a duel there in 1829.
Between 1850 and 1920, the Southwark and Vauxhall Water Company used the site as a water treatment and storage area. From this time the area was split between the Great Western Railway, which used it as a goods yard, and the London Power Company.
The coal-fired Battersea Power Station, built in two phases between 1929 and 1953, is essentially two separate facilities forming a near mirror image at the centre of the building.
Battersea A was built between 1929 and 1932. Diversion of labour to help the Second World War effort meant completion of Battersea B was delayed until 1953.
Each station contained a boiler house, a turbine hall and ancillary switchrooms and transformers. Below ground were a series of cooling water and power tunnels.
A desk study was the first step in the investigation. Archive site and aerial photographs, reports on construction and the archive of design and construction drawings helped in assessing the foundations for reuse.
This material provided a valuable insight into construction of 75% of the foundations, but it did not give the whole picture.
The study established that Battersea A was built within a sheet pile cofferdam through a cover of made ground, alluvium and terrace gravel and cut off within the London Clay at about 10m below site level.
During this initial work two deep closely connected drift-filled hollows were encountered through the near surface horizons, increasing the depth to London Clay to 33m below ground within a local area. The drift infill comprises a mixture of poorly consolidated sands, silts, gravels, peat and reworked London Clay.
The provenance of this infill, purported to be London's deepest 'scour' feature, has been the subject of much debate. Theories include a drift infill to a sink-hole within the underlying Upper Chalk or an erosional scour from the confluence of periglacial meltwaters. Ground investigations have not been able to support either theory.
An alternative is offered by the growth of ice bodies or pingos, which are believed to have been common across periglacial London and southern Britain. In any case the presence of this deep drift feature will continue to play a major role in the design of foundations for the station.
Given that the power station incorporated a single level of basement at +0.6m OD, foundation design without the 'scour' was based on spread footings into the London Clay, assuming the clay was encountered at an equivalent depth of less than -6m OD.
The distribution of the 'scour' complicated this issue and led to the majority of foundations for A station being supported on 430mm diameter vibro piles driven into unworked London Clay.
Piles were not always used, as rafts and other forms of spread foundations were incorporated in areas of shallow London Clay or where the form of construction, such as the cooling water inlet suction chamber and pump house, necessitated deeper foundations. The majority of vibro piles were driven to depths of less than 14m below basement level.
Surprisingly, fewer records were available for the B Station.
The works were carried out within a sheet pile cofferdam, but for some unknown reason were developed on a structural grid pattern that was offset to the initial building.
Where available the records indicated the prevalence of shallow pad foundations within the London Clay at depths of less than -6m OD and consistent with a geological profile away from the scour. Although piled foundations were again selected when London Clay was not encountered at shallow depths, an alternative solution involving pad foundations and chemical grouting of the terrace gravels above the London Clay was adopted in a number of cases.
The extensive legacy of buried foundations and tunnels is a potentially serious constraint to redevelopment. However, the team took the opposite view, recognising the potential for reusing the foundations, which in most cases have previously taken a greater load than that envisaged for the new build.
A number of challenges need to be met before the foundations can be considered for any new project, not least of which is demonstrating their value to project insurers and funders.
As second or third generation projects become more common, Buro Happold has begun to develop procedures to evaluate project suitability during the initial feasibility stage. Even if a project can be identified as appropriate, it is still necessary to carry out a vigorous ground investigation to support the idea.
In the absence of UK guidance, Buro Happold developed a number of procedures for the Battersea Power Station project based on several schemes, including the Millennium Dome, where foundation reuse had been considered.
As a starting point and to provide precise locations on which to focus investigations into foundation reuse, a time domain based electromagnetic (EM) and ground probing radar (GPR) survey was undertaken in the accessible areas of the building. This was carried out by Aperio acting as subcontractor to Geotechnics, the overall site investigation contractor, which also employed Technotrade for the investigation work.
This survey concentrated on the former boiler houses of A and B stations and also the turbine hall to A station where the bulk of any new build will take place. Access to B station is currently restricted on health and safety grounds.
The results of the EM survey were particularly useful in mapping out the locations of buried foundations and were even able to identify areas where foundations had been removed or modified during an earlier attempt at redevelopment in the late 1980s.
The use of geophysics as a nonintrusive method of investigation was a central theme of this study and was chosen to avoid excessive damage to the existing structures.
In addition to GPR and EM surveying, sonic echo and parallel seismic testing was undertaken at a number of foundations.
The former was used to verify integrity of piles and the latter to establish pile length and hence indirectly evaluate pile load capacity. Confirming pile length was critical to the investigation given the likely impact of the scour on pile toe elevations.
Despite the emphasis on the geophysics, it was still necessary to carry out some intrusive investigation. Within the power station this included a number of cable percussion boreholes to further delineate the scour and rotarydrilled wireline boreholes to determine the geological profile of the Upper Chalk for any new deep foundations.
Self-boring pressuremeter testing of the London Clay Formation and Thanet Sands was also undertaken to address any potential deep foundations solution. Machine dug trial pits were required to expose the foundation bases and verify the construction details recorded by the desk study.
Principally these were pile, pile cap and pad dimensions. The pits provided a visual assessment of their condition and were used to collect soil and groundwater samples in order to identify potentially aggressive materials near the foundations as well as the exact location of subsequent parallel seismic and integrity testing.
Both forms of geophysical test were carried out on piles that remained integral to the cap to avoid unnecessary damage. The parallel seismic test involves striking a pile head or pad foundation and measuring the travel time of the resulting compressive stress wave through the foundation into a receiving hydrophone, which is lowered in progressive increments through an adjacent borehole.
The foundation base is taken when there is a change of gradient in the travel time/depth plot.
As a secondary measure foundation depths were also estimated from the sonic echo integrity tests, but the pile dimensions can limit the effectiveness of this method as the receiving signal needs to be reflected through the test pile.
Parallel seismic tests were undertaken to verify the base of pad foundations. The results of testing are under review but the field data suggests that in most cases it was possible to confirm the predicted foundation dimensions and establish an estimate of their capacity.
To calibrate the geophysical tests or in cases where a clear toe response was not determined, rotary probe holes were carried out. This provided further insight in some examples but was not always entirely successful because of the tight tolerances on verticality.
To verify the field observations on concrete and reinforcement quality, a series of vertical and horizontal cores were taken through the piles, pile caps and pad foundations to confirm physical and chemical properties in the laboratory.
Fieldwork has just been completed. After laboratory and site data have been evaluated, detailed design will examine solutions for reuse of the foundations.
Justin Phillips is an associate of Buro Happold Ground Engineering and is project leader for the ground investigation at Battersea.