SITE INVESTIGATION and routine foundation design could change fundamentally if industry adopts an approach proposed by Professor John Atkinson during his delivery of the 40th Rankine Lecture in London last month.
Professor Atkinson of City University used the prestigious Rankine platform to describe how engineers could account for soil nonlinearity in routine design.
'We like to believe that engineering properties of soils are variable. Indeed they are but they do vary in a consistent and predictable way. If they did not soil mechanics would be witchcraft, 'he said.
Atkinson explained to the 700-strong audience how stiffness is an important parameter for characterising non-linearity and so determining design.
He showed that settlement of foundations both in the ground and in model tests is influenced by non-linear soil behaviour.
The ability to measure stiffness at very small strain is key to the approach, but this no longer requires use of complicated research level laboratory testing facilities, he said.
Relatively simple routine tests, for example laboratory bender element tests and insitu shear wave velocity tests, can measure soil stiffness at very small strain.
Atkinson also outlined a procedure for routine design of simple structures incorporating non-linear soil stiffness.
In closing he issued a challenge to equipment designers and manufacturers to further develop methods for measuring these parameters.
'Techniques for measuring shear wave velocity to determine very small strain stiffness are well established but equipment needs development to make it more robust and user-friendly.'
In the vote of thanks former Rankine Lecturer Brian Simpson of Ove Arup said: 'John has shown how the understanding developed in modern laboratory testing can be applied to useful effect by practitioners.
'He has reduced the complexity to a small number of parameters and has shown how settlements may be assessed for the full range of loading from zero to failure on the basis of the stiffness-strain relationship.
'This is a problem that has always confronted designers of foundations. We are better at estimating ultimate failure, which is generally a remote and almost irrelevant possibility, than we are at assessing settlement in service.'
Atkinson's paper will be published in Geotechnique.
l The Rankine Dinner was attended by 400 of the audience, including leading geotechnical engineers from the UK and around the world.
Guests included Heinz Brandl, professor of geotechnical engineering at Vienna Technical University in Austria, who will be next year's Rankine Lecturer.Brandl is also ISSMGE's vice president for Europe.A raffle on behalf of RedR raised £1850.