During the shaft sinking works for the London Power Tunnels National Grid carried out settlement measurements around shafts varying in outside diameter from 6.5m to a maximum of 16.4m. This note presents a summary of this data and suggests a simple relationship for settlement prediction for shafts of varying diameters in stiff cohesive materials.
In 1994 New and Bowers published measurements of ground movements carried out during the works for the Heathrow Express Trial tunnel. These measurements included settlements around an 11m OD, 26m deep, segmentally-lined shaft into London Clay overlain with Terrace Gravels. Their equation provided a good fit with the Heathrow data but did of course rely on data from just one shaft diameter.
Since publication this equation has been widely used (and abused) and has been corroborated by measurements taken elsewhere for similar sized shafts in stiff clays. This technical note seeks to use the National Grid data to extend the predictive relevance of the relationship to shafts of varying diameters.
THE NATIONAL GRID DATA
Figure 1 shows the settlement data from 13 shafts. The shafts were of similar construction as caisson segmental with sprayed concrete underpinning as required. Table 1 shows the location, size and depths of the shafts which were all in the London area and predominantly into stiff cohesive materials overlain by sands and gravels. The data has been assembled and synthesised as part of a wider study on circular shaft construction by Faustin (2017).
The data shown in blue were from the largest four shafts which were nominally 15.7m OD. Data in green were from eight shafts of nominally 11.1m and 13.2m OD and these are grouped together as there was no clear difference between them. The red data were from a single shaft of nominally 6.5m OD. The depths of the shafts varied between 26.5m and 47m. The data shown as colourless points are from the two shafts at Kensal Green (13.2m and 15.7m OD) which showed what appeared to be anomalously low settlements: At present the reason for this is uncertain. There is a substantial scatter in the data probably arising from the normal difficulties in obtaining data in urban environments close to major excavation works. In particular, the sites around shafts are usually heavily impacted by construction plant and other traffic and the measuring points are often mounted on pavement structures of varying stability. The wide spread in the data makes analysis difficult but there is a clear trend that larger shafts tend, as might be expected, to produce larger settlements of greater extent.
There is currently insufficient field data to reliably predict horizontal movements which are commonly assumed to occur radially toward the shaft and to be of similar magnitude to the settlement.
CONCLUSIONS
The suggested generic equations may be considered by designers as predictive tools and the values of ?? and n chosen so as to reflect the degree of conservatism required of their assessment (i.e. conservative, moderately conservative, or best estimate): The values chosen should be supported by case history data suited to the shaft under assessment.
It must be noted that the data given in this note are derived from stiff cohesive materials overlain by various thicknesses of made ground and drift deposits (mainly sands and gravels). Elsewhere very different ground conditions will produce different results although the same equation may still be helpful. For instance, in Sao Paulo significantly greater and more extensive settlements (e.g. ?? = 0.23 per cent and n = 1.8) have been indicated particularly in very soft to medium clays and where local dewatering had taken place (Dias et al, 2015).
There is an increasing use of diaphragm walls particularly for larger diameter shafts and such developments are hoped to reduce settlements: However at present there is a very limited database and predictions are varied and difficult and more reliable data, particularly during wall installation, are required.
Where settlement is an important issue the wide scatter of results may be reduced if shallow ground anchors were used as a supplement (or replacement) to surface mounted road studs. A sketch of an anchor design as used at Heathrow and on other current major projects in London is given in Figure 2.
